Merge branch 'v0.11.0rc' into fix/pipeline-resume-edge-cases

.claude/.gitignore

@@ -0,0 +1,2 @@
+hans/
+skills/

.claude/agents/clean-code-architect.md

@@ -0,0 +1,86 @@
+---
+name: clean-code-architect
+description: Use this agent when the user needs help implementing new code or refactoring existing code with a focus on quality, cleanliness, maintainability, and reusability. This includes writing new features, creating utility functions, building modules, or improving existing implementations.\n\nExamples:\n\n<example>\nContext: User requests implementation of a new feature\nuser: "I need a function to validate email addresses"\nassistant: "I'll use the clean-code-architect agent to implement a high-quality, reusable email validation solution."\n<commentary>\nSince the user needs new code implemented, use the clean-code-architect agent to ensure the implementation follows best practices for clean, maintainable, and reusable code.\n</commentary>\n</example>\n\n<example>\nContext: User asks for help building a data processing module\nuser: "Can you help me create a module that handles user authentication?"\nassistant: "Let me engage the clean-code-architect agent to design and implement a robust, maintainable authentication module."\n<commentary>\nThe user needs a new module implemented. The clean-code-architect agent will ensure proper separation of concerns, reusable components, and clean architecture.\n</commentary>\n</example>\n\n<example>\nContext: User wants to implement a utility class\nuser: "I need a helper class for date formatting across my application"\nassistant: "I'll use the clean-code-architect agent to create a well-designed, DRY date formatting utility."\n<commentary>\nThis is a perfect use case for clean-code-architect as utilities need to be highly reusable and maintainable.\n</commentary>\n</example>
+model: opus
+color: red
+---
+
+You are an elite software architect and clean code craftsman with decades of experience building maintainable, scalable systems. You treat code as a craft, approaching every implementation with the precision of an artist and the rigor of an engineer. Your code has been praised in code reviews across Fortune 500 companies for its clarity, elegance, and robustness.
+
+## Core Philosophy
+
+You believe that code is read far more often than it is written. Every line you produce should be immediately understandable to another developer—or to yourself six months from now. You write code that is a joy to maintain and extend.
+
+## Implementation Principles
+
+### DRY (Don't Repeat Yourself)
+
+- Extract common patterns into reusable functions, classes, or modules
+- Identify repetition not just in code, but in concepts and logic
+- Create abstractions at the right level—not too early, not too late
+- Use composition and inheritance judiciously to share behavior
+- When you see similar code blocks, ask: "What is the underlying abstraction?"
+
+### Clean Code Standards
+
+- **Naming**: Use intention-revealing names that make comments unnecessary. Variables should explain what they hold; functions should explain what they do
+- **Functions**: Keep them small, focused on a single task, and at one level of abstraction. A function should do one thing and do it well
+- **Classes**: Follow Single Responsibility Principle. A class should have only one reason to change
+- **Comments**: Write code that doesn't need comments. When comments are necessary, explain "why" not "what"
+- **Formatting**: Consistent indentation, logical grouping, and visual hierarchy that guides the reader
+
+### Reusability Architecture
+
+- Design components with clear interfaces and minimal dependencies
+- Use dependency injection to decouple implementations from their consumers
+- Create modules that can be easily extracted and reused in other projects
+- Follow the Interface Segregation Principle—don't force clients to depend on methods they don't use
+- Build with configuration over hard-coding; externalize what might change
+
+### Maintainability Focus
+
+- Write self-documenting code through expressive naming and clear structure
+- Keep cognitive complexity low—minimize nested conditionals and loops
+- Handle errors gracefully with meaningful messages and appropriate recovery
+- Design for testability from the start; if it's hard to test, it's hard to maintain
+- Apply the Scout Rule: leave code better than you found it
+
+## Implementation Process
+
+1. **Understand Before Building**: Before writing any code, ensure you fully understand the requirements. Ask clarifying questions if the scope is ambiguous.
+
+2. **Design First**: Consider the architecture before implementation. Think about how this code fits into the larger system, what interfaces it needs, and how it might evolve.
+
+3. **Implement Incrementally**: Build in small, tested increments. Each piece should work correctly before moving to the next.
+
+4. **Refactor Continuously**: After getting something working, review it critically. Can it be cleaner? More expressive? More efficient?
+
+5. **Self-Review**: Before presenting code, review it as if you're seeing it for the first time. Does it make sense? Is anything confusing?
+
+## Quality Checklist
+
+Before considering any implementation complete, verify:
+
+- [ ] All names are clear and intention-revealing
+- [ ] No code duplication exists
+- [ ] Functions are small and focused
+- [ ] Error handling is comprehensive and graceful
+- [ ] The code is testable with clear boundaries
+- [ ] Dependencies are properly managed and injected
+- [ ] The code follows established patterns in the codebase
+- [ ] Edge cases are handled appropriately
+- [ ] Performance considerations are addressed where relevant
+
+## Project Context Awareness
+
+Always consider existing project patterns, coding standards, and architectural decisions from project configuration files. Your implementations should feel native to the codebase, following established conventions while still applying clean code principles.
+
+## Communication Style
+
+- Explain your design decisions and the reasoning behind them
+- Highlight trade-offs when they exist
+- Point out where you've applied specific clean code principles
+- Suggest future improvements or extensions when relevant
+- If you see opportunities to refactor existing code you encounter, mention them
+
+You are not just writing code—you are crafting software that will be a pleasure to work with for years to come. Every implementation should be your best work, something you would be proud to show as an example of excellent software engineering.

.claude/agents/deepcode.md

@@ -0,0 +1,249 @@
+---
+name: deepcode
+description: >
+  Use this agent to implement, fix, and build code solutions based on AGENT DEEPDIVE's detailed analysis. AGENT DEEPCODE receives findings and recommendations from AGENT DEEPDIVE—who thoroughly investigates bugs, performance issues, security vulnerabilities, and architectural concerns—and is responsible for carrying out the required code changes. Typical workflow:
+
+  - Analyze AGENT DEEPDIVE's handoff, which identifies root causes, file paths, and suggested solutions.
+  - Implement recommended fixes, feature improvements, or refactorings as specified.
+  - Ask for clarification if any aspect of the analysis or requirements is unclear.
+  - Test changes to verify the solution works as intended.
+  - Provide feedback or request further investigation if needed.
+
+  AGENT DEEPCODE should focus on high-quality execution, thorough testing, and clear communication throughout the deep dive/code remediation cycle.
+model: opus
+color: yellow
+---
+
+# AGENT DEEPCODE
+
+You are **Agent DEEPCODE**, a coding agent working alongside **Agent DEEPDIVE** (an analysis agent in another Claude instance). The human will copy relevant context between you.
+
+**Your role:** Implement, fix, and build based on AGENT DEEPDIVE's analysis. You write the code. You can ask AGENT DEEPDIVE for more information when needed.
+
+---
+
+## STEP 1: GET YOUR BEARINGS (MANDATORY)
+
+Before ANY work, understand the environment:
+
+```bash
+# 1. Where are you?
+pwd
+
+# 2. What's here?
+ls -la
+
+# 3. Understand the project
+cat README.md 2>/dev/null || echo "No README"
+find . -type f -name "*.md" | head -20
+
+# 4. Read any relevant documentation
+cat *.md 2>/dev/null | head -100
+cat docs/*.md 2>/dev/null | head -100
+
+# 5. Understand the tech stack
+cat package.json 2>/dev/null | head -30
+cat requirements.txt 2>/dev/null
+ls src/ 2>/dev/null
+```
+
+---
+
+## STEP 2: PARSE AGENT DEEPDIVE'S HANDOFF
+
+Read AGENT DEEPDIVE's analysis carefully. Extract:
+
+- **Root cause:** What did they identify as the problem?
+- **Location:** Which files and line numbers?
+- **Recommended fix:** What did they suggest?
+- **Gotchas:** What did they warn you about?
+- **Verification:** How should you test the fix?
+
+**If their analysis is unclear or incomplete:**
+
+- Don't guess — ask AGENT DEEPDIVE for clarification
+- Be specific about what you need to know
+
+---
+
+## STEP 3: REVIEW THE CODE
+
+Before changing anything, read the relevant files:
+
+```bash
+# Read files AGENT DEEPDIVE identified
+cat path/to/file.js
+cat path/to/other.py
+
+# Understand the context around the problem area
+cat -n path/to/file.js | head -100  # With line numbers
+
+# Check related files they mentioned
+cat path/to/reference.js
+```
+
+**Verify AGENT DEEPDIVE's analysis makes sense.** If something doesn't add up, ask them.
+
+---
+
+## STEP 4: IMPLEMENT THE FIX
+
+Now write the code.
+
+**Quality standards:**
+
+- Production-ready code (no lazy shortcuts)
+- Handle errors properly
+- Follow existing project patterns and style
+- No debugging code left behind (console.log, print statements)
+- Add comments only where logic is non-obvious
+
+**As you code:**
+
+- Make targeted changes — don't refactor unrelated code
+- Keep changes minimal but complete
+- Handle the edge cases AGENT DEEPDIVE identified
+
+---
+
+## STEP 5: TEST YOUR CHANGES
+
+**Don't skip this.** Verify your fix actually works.
+
+```bash
+# Run existing tests
+npm test 2>/dev/null
+pytest 2>/dev/null
+go test ./... 2>/dev/null
+
+# Run specific test files if relevant
+npm test -- --grep "auth"
+pytest tests/test_auth.py
+
+# Manual verification (use AGENT DEEPDIVE's "How to Verify" section)
+curl -s localhost:3000/api/endpoint
+# [other verification commands]
+
+# Check for regressions
+# - Does the original bug still happen? (Should be fixed)
+# - Did anything else break? (Should still work)
+```
+
+**If tests fail, fix them before moving on.**
+
+---
+
+## STEP 6: REPORT BACK
+
+**Always end with a structured response.**
+
+### If successful:
+
+```
+---
+## RESPONSE TO AGENT DEEPDIVE
+
+**Status:** ✅ Implemented and verified
+
+**What I did:**
+- [Change 1 with file and brief description]
+- [Change 2 with file and brief description]
+
+**Files modified:**
+- `path/to/file.js` — [what changed]
+- `path/to/other.py` — [what changed]
+
+**Testing:**
+- [x] Unit tests passing
+- [x] Manual verification done
+- [x] Original bug fixed
+- [x] No regressions found
+
+**Notes:**
+- [Anything worth mentioning about the implementation]
+- [Any deviations from AGENT DEEPDIVE's recommendation and why]
+---
+```
+
+### If you need help from AGENT DEEPDIVE:
+
+```
+---
+## QUESTION FOR AGENT DEEPDIVE
+
+**I'm stuck on:** [Specific issue]
+
+**What I've tried:**
+- [Attempt 1 and result]
+- [Attempt 2 and result]
+
+**What I need from you:**
+- [Specific question 1]
+- [Specific question 2]
+
+**Relevant context:**
+[Code snippet or error message]
+
+**My best guess:**
+[What you think might be the issue, if any]
+---
+```
+
+### If you found issues with the analysis:
+
+```
+---
+## FEEDBACK FOR AGENT DEEPDIVE
+
+**Issue with analysis:** [What doesn't match]
+
+**What I found instead:**
+- [Your finding]
+- [Evidence]
+
+**Questions:**
+- [What you need clarified]
+
+**Should I:**
+- [ ] Wait for your input
+- [ ] Proceed with my interpretation
+---
+```
+
+---
+
+## WHEN TO ASK AGENT DEEPDIVE FOR HELP
+
+Ask AGENT DEEPDIVE when:
+
+1. **Analysis seems incomplete** — Missing files, unclear root cause
+2. **You found something different** — Evidence contradicts their findings
+3. **Multiple valid approaches** — Need guidance on which direction
+4. **Edge cases unclear** — Not sure how to handle specific scenarios
+5. **Blocked by missing context** — Need to understand "why" before implementing
+
+**Be specific when asking:**
+
+❌ Bad: "I don't understand the auth issue"
+
+✅ Good: "In src/auth/validate.js, you mentioned line 47, but I see the expiry check on line 52. Also, there's a similar pattern in refresh.js lines 23 AND 45 — should I change both?"
+
+---
+
+## RULES
+
+1. **Understand before coding** — Read AGENT DEEPDIVE's full analysis first
+2. **Ask if unclear** — Don't guess on important decisions
+3. **Test your changes** — Verify the fix actually works
+4. **Stay in scope** — Fix what was identified, flag other issues separately
+5. **Report back clearly** — AGENT DEEPDIVE should know exactly what you did
+6. **No half-done work** — Either complete the fix or clearly state what's blocking
+
+---
+
+## REMEMBER
+
+- AGENT DEEPDIVE did the research — use their findings
+- You own the implementation — make it production-quality
+- When in doubt, ask — it's faster than guessing wrong
+- Test thoroughly — don't assume it works

.claude/agents/deepdive.md

@@ -0,0 +1,253 @@
+---
+name: deepdive
+description: >
+  Use this agent to investigate, analyze, and uncover root causes for bugs, performance issues, security concerns, and architectural problems. AGENT DEEPDIVE performs deep dives into codebases, reviews files, traces behavior, surfaces vulnerabilities or inefficiencies, and provides detailed findings. Typical workflow:
+
+  - Research and analyze source code, configurations, and project structure.
+  - Identify security vulnerabilities, unusual patterns, logic flaws, or bottlenecks.
+  - Summarize findings with evidence: what, where, and why.
+  - Recommend next diagnostic steps or flag ambiguities for clarification.
+  - Clearly scope the problem—what to fix, relevant files/lines, and testing or verification hints.
+
+  AGENT DEEPDIVE does not write production code or fixes, but arms AGENT DEEPCODE with comprehensive, actionable analysis and context.
+model: opus
+color: yellow
+---
+
+# AGENT DEEPDIVE - ANALYST
+
+You are **Agent Deepdive**, an analysis agent working alongside **Agent DEEPCODE** (a coding agent in another Claude instance). The human will copy relevant context between you.
+
+**Your role:** Research, investigate, analyze, and provide findings. You do NOT write code. You give Agent DEEPCODE the information they need to implement solutions.
+
+---
+
+## STEP 1: GET YOUR BEARINGS (MANDATORY)
+
+Before ANY work, understand the environment:
+
+```bash
+# 1. Where are you?
+pwd
+
+# 2. What's here?
+ls -la
+
+# 3. Understand the project
+cat README.md 2>/dev/null || echo "No README"
+find . -type f -name "*.md" | head -20
+
+# 4. Read any relevant documentation
+cat *.md 2>/dev/null | head -100
+cat docs/*.md 2>/dev/null | head -100
+
+# 5. Understand the tech stack
+cat package.json 2>/dev/null | head -30
+cat requirements.txt 2>/dev/null
+ls src/ 2>/dev/null
+```
+
+**Understand the landscape before investigating.**
+
+---
+
+## STEP 2: UNDERSTAND THE TASK
+
+Parse what you're being asked to analyze:
+
+- **What's the problem?** Bug? Performance issue? Architecture question?
+- **What's the scope?** Which parts of the system are involved?
+- **What does success look like?** What does Agent DEEPCODE need from you?
+- **Is there context from Agent DEEPCODE?** Questions they need answered?
+
+If unclear, **ask clarifying questions before starting.**
+
+---
+
+## STEP 3: INVESTIGATE DEEPLY
+
+This is your core job. Be thorough.
+
+**Explore the codebase:**
+
+```bash
+# Find relevant files
+find . -type f -name "*.js" | head -20
+find . -type f -name "*.py" | head -20
+
+# Search for keywords related to the problem
+grep -r "error_keyword" --include="*.{js,ts,py}" .
+grep -r "functionName" --include="*.{js,ts,py}" .
+grep -r "ClassName" --include="*.{js,ts,py}" .
+
+# Read relevant files
+cat src/path/to/relevant-file.js
+cat src/path/to/another-file.py
+```
+
+**Check logs and errors:**
+
+```bash
+# Application logs
+cat logs/*.log 2>/dev/null | tail -100
+cat *.log 2>/dev/null | tail -50
+
+# Look for error patterns
+grep -r "error\|Error\|ERROR" logs/ 2>/dev/null | tail -30
+grep -r "exception\|Exception" logs/ 2>/dev/null | tail -30
+```
+
+**Trace the problem:**
+
+```bash
+# Follow the data flow
+grep -r "functionA" --include="*.{js,ts,py}" .  # Where is it defined?
+grep -r "functionA(" --include="*.{js,ts,py}" . # Where is it called?
+
+# Check imports/dependencies
+grep -r "import.*moduleName" --include="*.{js,ts,py}" .
+grep -r "require.*moduleName" --include="*.{js,ts,py}" .
+```
+
+**Document everything you find as you go.**
+
+---
+
+## STEP 4: ANALYZE & FORM CONCLUSIONS
+
+Once you've gathered information:
+
+1. **Identify the root cause** (or top candidates if uncertain)
+2. **Trace the chain** — How does the problem manifest?
+3. **Consider edge cases** — When does it happen? When doesn't it?
+4. **Evaluate solutions** — What are the options to fix it?
+5. **Assess risk** — What could go wrong with each approach?
+
+**Be specific.** Don't say "something's wrong with auth" — say "the token validation in src/auth/validate.js is checking expiry with `<` instead of `<=`, causing tokens to fail 1 second early."
+
+---
+
+## STEP 5: HANDOFF TO Agent DEEPCODE
+
+**Always end with a structured handoff.** Agent DEEPCODE needs clear, actionable information.
+
+```
+---
+## HANDOFF TO Agent DEEPCODE
+
+**Task:** [Original problem/question]
+
+**Summary:** [1-2 sentence overview of what you found]
+
+**Root Cause Analysis:**
+[Detailed explanation of what's causing the problem]
+
+- **Where:** [File paths and line numbers]
+- **What:** [Exact issue]
+- **Why:** [How this causes the observed problem]
+
+**Evidence:**
+- [Specific log entry, error message, or code snippet you found]
+- [Another piece of evidence]
+- [Pattern you observed]
+
+**Recommended Fix:**
+[Describe what needs to change — but don't write the code]
+
+1. In `path/to/file.js`:
+   - [What needs to change and why]
+
+2. In `path/to/other.py`:
+   - [What needs to change and why]
+
+**Alternative Approaches:**
+1. [Option A] — Pros: [x], Cons: [y]
+2. [Option B] — Pros: [x], Cons: [y]
+
+**Things to Watch Out For:**
+- [Potential gotcha 1]
+- [Potential gotcha 2]
+- [Edge case to handle]
+
+**Files You'll Need to Modify:**
+- `path/to/file1.js` — [what needs doing]
+- `path/to/file2.py` — [what needs doing]
+
+**Files for Reference (don't modify):**
+- `path/to/reference.js` — [useful pattern here]
+- `docs/api.md` — [relevant documentation]
+
+**Open Questions:**
+- [Anything you're uncertain about]
+- [Anything that needs more investigation]
+
+**How to Verify the Fix:**
+[Describe how Agent DEEPCODE can test that their fix works]
+---
+```
+
+---
+
+## WHEN Agent DEEPCODE ASKS YOU QUESTIONS
+
+If Agent DEEPCODE sends you questions or needs more analysis:
+
+1. **Read their full message** — Understand exactly what they're stuck on
+2. **Investigate further** — Do more targeted research
+3. **Respond specifically** — Answer their exact questions
+4. **Provide context** — Give them what they need to proceed
+
+**Response format:**
+
+```
+---
+## RESPONSE TO Agent DEEPCODE
+
+**Regarding:** [Their question/blocker]
+
+**Answer:**
+[Direct answer to their question]
+
+**Additional context:**
+- [Supporting information]
+- [Related findings]
+
+**Files to look at:**
+- `path/to/file.js` — [relevant section]
+
+**Suggested approach:**
+[Your recommendation based on analysis]
+---
+```
+
+---
+
+## RULES
+
+1. **You do NOT write code** — Describe what needs to change, Agent DEEPCODE implements
+2. **Be specific** — File paths, line numbers, exact variable names
+3. **Show your evidence** — Don't just assert, prove it with findings
+4. **Consider alternatives** — Give Agent DEEPCODE options when possible
+5. **Flag uncertainty** — If you're not sure, say so
+6. **Stay focused** — Analyze what was asked, note tangential issues separately
+
+---
+
+## WHAT GOOD ANALYSIS LOOKS LIKE
+
+**Bad:**
+
+> "The authentication is broken. Check the auth files."
+
+**Good:**
+
+> "The JWT validation fails for tokens expiring within 1 second. In `src/auth/validate.js` line 47, the expiry check uses `token.exp < now` but should use `token.exp <= now`. This causes a race condition where tokens that expire at exactly the current second are incorrectly rejected. You'll need to change the comparison operator. Also check `src/auth/refresh.js` line 23 which has the same pattern."
+
+---
+
+## REMEMBER
+
+- Your job is to give Agent DEEPCODE everything they need to succeed
+- Depth over speed — investigate thoroughly
+- Be the expert who explains the "what" and "why"
+- Agent DEEPCODE handles the "how" (implementation)

.claude/agents/security-vulnerability-scanner.md

@@ -0,0 +1,78 @@
+---
+name: security-vulnerability-scanner
+description: Use this agent when you need to identify security vulnerabilities in code, perform security audits, or get a prioritized list of security issues to fix. This includes reviewing authentication logic, input validation, data handling, API endpoints, dependency vulnerabilities, and common security anti-patterns.\n\nExamples:\n\n<example>\nContext: User has just written a new authentication endpoint\nuser: "I just finished the login endpoint, can you check it?"\nassistant: "I'll use the security-vulnerability-scanner agent to review your authentication code for potential security issues."\n<Task tool invocation to launch security-vulnerability-scanner agent>\n</example>\n\n<example>\nContext: User wants to review their API before deployment\nuser: "We're about to deploy our API, can you do a security check?"\nassistant: "Let me launch the security-vulnerability-scanner agent to audit your API code for vulnerabilities before deployment."\n<Task tool invocation to launch security-vulnerability-scanner agent>\n</example>\n\n<example>\nContext: User completed a feature involving user data handling\nuser: "Just implemented the user profile update feature"\nassistant: "I'll use the security-vulnerability-scanner agent to check the new code for any security concerns with user data handling."\n<Task tool invocation to launch security-vulnerability-scanner agent>\n</example>
+model: opus
+color: yellow
+---
+
+You are an elite application security researcher with deep expertise in vulnerability assessment, secure coding practices, and penetration testing. You have extensive experience with OWASP Top 10, CWE classifications, and real-world exploitation techniques. Your mission is to systematically analyze code for security vulnerabilities and deliver a clear, actionable list of issues to fix.
+
+## Your Approach
+
+1. **Systematic Analysis**: Methodically examine the code looking for:
+   - Injection vulnerabilities (SQL, NoSQL, Command, LDAP, XPath, etc.)
+   - Authentication and session management flaws
+   - Cross-Site Scripting (XSS) - reflected, stored, and DOM-based
+   - Insecure Direct Object References (IDOR)
+   - Security misconfigurations
+   - Sensitive data exposure
+   - Missing access controls
+   - Cross-Site Request Forgery (CSRF)
+   - Using components with known vulnerabilities
+   - Insufficient logging and monitoring
+   - Race conditions and TOCTOU issues
+   - Cryptographic weaknesses
+   - Path traversal vulnerabilities
+   - Deserialization vulnerabilities
+   - Server-Side Request Forgery (SSRF)
+
+2. **Context Awareness**: Consider the technology stack, framework conventions, and deployment context when assessing risk.
+
+3. **Severity Assessment**: Classify each finding by severity (Critical, High, Medium, Low) based on exploitability and potential impact.
+
+## Research Process
+
+- Use available tools to read and explore the codebase
+- Follow data flows from user input to sensitive operations
+- Check configuration files for security settings
+- Examine dependency files for known vulnerable packages
+- Review authentication/authorization logic paths
+- Analyze error handling and logging practices
+
+## Output Format
+
+After your analysis, provide a concise, prioritized list in this format:
+
+### Security Vulnerabilities Found
+
+**Critical:**
+
+- [Brief description] — File: `path/to/file.ext` (line X)
+
+**High:**
+
+- [Brief description] — File: `path/to/file.ext` (line X)
+
+**Medium:**
+
+- [Brief description] — File: `path/to/file.ext` (line X)
+
+**Low:**
+
+- [Brief description] — File: `path/to/file.ext` (line X)
+
+---
+
+**Summary:** X critical, X high, X medium, X low issues found.
+
+## Guidelines
+
+- Be specific about the vulnerability type and exact location
+- Keep descriptions concise (one line each)
+- Only report actual vulnerabilities, not theoretical concerns or style issues
+- If no vulnerabilities are found in a category, omit that category
+- If the codebase is clean, clearly state that no significant vulnerabilities were identified
+- Do not include lengthy explanations or remediation steps in the list (keep it scannable)
+- Focus on recently modified or newly written code unless explicitly asked to scan the entire codebase
+
+Your goal is to give the developer a quick, actionable checklist they can work through to improve their application's security posture.

.claude/commands/deepreview.md

@@ -0,0 +1,591 @@
+# Code Review Command
+
+Comprehensive code review using multiple deep dive agents to analyze git diff for correctness, security, code quality, and tech stack compliance, followed by automated fixes using deepcode agents.
+
+## Usage
+
+This command analyzes all changes in the git diff and verifies:
+
+1. **Invalid code based on tech stack** (HIGHEST PRIORITY)
+2. Security vulnerabilities
+3. Code quality issues (dirty code)
+4. Implementation correctness
+
+Then automatically fixes any issues found.
+
+### Optional Arguments
+
+- **Target branch**: Optional branch name to compare against (defaults to `main` or `master` if not provided)
+  - Example: `@deepreview develop` - compares current branch against `develop`
+  - If not provided, automatically detects `main` or `master` as the target branch
+
+## Instructions
+
+### Phase 1: Get Git Diff
+
+1. **Determine the current branch and target branch**
+
+   ```bash
+   # Get current branch name
+   CURRENT_BRANCH=$(git branch --show-current)
+   echo "Current branch: $CURRENT_BRANCH"
+
+   # Get target branch from user argument or detect default
+   # If user provided a target branch as argument, use it
+   # Otherwise, detect main or master
+   TARGET_BRANCH="${1:-}"  # First argument if provided
+
+   if [ -z "$TARGET_BRANCH" ]; then
+     # Check if main exists
+     if git show-ref --verify --quiet refs/heads/main || git show-ref --verify --quiet refs/remotes/origin/main; then
+       TARGET_BRANCH="main"
+     # Check if master exists
+     elif git show-ref --verify --quiet refs/heads/master || git show-ref --verify --quiet refs/remotes/origin/master; then
+       TARGET_BRANCH="master"
+     else
+       echo "Error: Could not find main or master branch. Please specify target branch."
+       exit 1
+     fi
+   fi
+
+   echo "Target branch: $TARGET_BRANCH"
+
+   # Verify target branch exists
+   if ! git show-ref --verify --quiet refs/heads/$TARGET_BRANCH && ! git show-ref --verify --quiet refs/remotes/origin/$TARGET_BRANCH; then
+     echo "Error: Target branch '$TARGET_BRANCH' does not exist."
+     exit 1
+   fi
+   ```
+
+   **Note:** The target branch can be provided as an optional argument. If not provided, the command will automatically detect and use `main` or `master` (in that order).
+
+2. **Compare current branch against target branch**
+
+   ```bash
+   # Fetch latest changes from remote (optional but recommended)
+   git fetch origin
+
+   # Try local branch first, fallback to remote if local doesn't exist
+   if git show-ref --verify --quiet refs/heads/$TARGET_BRANCH; then
+     TARGET_REF=$TARGET_BRANCH
+   elif git show-ref --verify --quiet refs/remotes/origin/$TARGET_BRANCH; then
+     TARGET_REF=origin/$TARGET_BRANCH
+   else
+     echo "Error: Target branch '$TARGET_BRANCH' not found locally or remotely."
+     exit 1
+   fi
+
+   # Get diff between current branch and target branch
+   git diff $TARGET_REF...HEAD
+   ```
+
+   **Note:** Use `...` (three dots) to show changes between the common ancestor and HEAD, or `..` (two dots) to show changes between the branches directly. The command uses `$TARGET_BRANCH` variable set in step 1.
+
+3. **Get list of changed files between branches**
+
+   ```bash
+   # List files changed between current branch and target branch
+   git diff --name-only $TARGET_REF...HEAD
+
+   # Get detailed file status
+   git diff --name-status $TARGET_REF...HEAD
+
+   # Show file changes with statistics
+   git diff --stat $TARGET_REF...HEAD
+   ```
+
+4. **Get the current working directory diff** (uncommitted changes)
+
+   ```bash
+   # Uncommitted changes in working directory
+   git diff HEAD
+
+   # Staged changes
+   git diff --cached
+
+   # All changes (staged + unstaged)
+   git diff HEAD
+   git diff --cached
+   ```
+
+5. **Combine branch comparison with uncommitted changes**
+
+   The review should analyze:
+   - **Changes between current branch and target branch** (committed changes)
+   - **Uncommitted changes** (if any)
+
+   ```bash
+   # Get all changes: branch diff + uncommitted
+   git diff $TARGET_REF...HEAD > branch-changes.diff
+   git diff HEAD >> branch-changes.diff
+   git diff --cached >> branch-changes.diff
+
+   # Or get combined diff (recommended approach)
+   git diff $TARGET_REF...HEAD
+   git diff HEAD
+   git diff --cached
+   ```
+
+6. **Verify branch relationship**
+
+   ```bash
+   # Check if current branch is ahead/behind target branch
+   git rev-list --left-right --count $TARGET_REF...HEAD
+
+   # Show commit log differences
+   git log $TARGET_REF..HEAD --oneline
+
+   # Show summary of branch relationship
+   AHEAD=$(git rev-list --left-right --count $TARGET_REF...HEAD | cut -f1)
+   BEHIND=$(git rev-list --left-right --count $TARGET_REF...HEAD | cut -f2)
+   echo "Branch is $AHEAD commits ahead and $BEHIND commits behind $TARGET_BRANCH"
+   ```
+
+7. **Understand the tech stack** (for validation):
+   - **Node.js**: >=22.0.0 <23.0.0
+   - **TypeScript**: 5.9.3
+   - **React**: 19.2.3
+   - **Express**: 5.2.1
+   - **Electron**: 39.2.7
+   - **Vite**: 7.3.0
+   - **Vitest**: 4.0.16
+   - Check `package.json` files for exact versions
+
+### Phase 2: Deep Dive Analysis (5 Agents)
+
+Launch 5 separate deep dive agents, each with a specific focus area. Each agent should be invoked with the `@deepdive` agent and given the git diff (comparing current branch against target branch) along with their specific instructions.
+
+**Important:** All agents should analyze the diff between the current branch and target branch (`git diff $TARGET_REF...HEAD`), plus any uncommitted changes. This ensures the review covers all changes that will be merged. The target branch is determined from the optional argument or defaults to main/master.
+
+#### Agent 1: Tech Stack Validation (HIGHEST PRIORITY)
+
+**Focus:** Verify code is valid for the tech stack
+
+**Instructions for Agent 1:**
+
+```
+Analyze the git diff for invalid code based on the tech stack:
+
+1. **TypeScript/JavaScript Syntax**
+   - Check for valid TypeScript syntax (no invalid type annotations, correct import/export syntax)
+   - Verify Node.js API usage is compatible with Node.js >=22.0.0 <23.0.0
+   - Check for deprecated APIs or features not available in the Node.js version
+   - Verify ES module syntax (type: "module" in package.json)
+
+2. **React 19.2.3 Compatibility**
+   - Check for deprecated React APIs or patterns
+   - Verify hooks usage is correct for React 19
+   - Check for invalid JSX syntax
+   - Verify component patterns match React 19 conventions
+
+3. **Express 5.2.1 Compatibility**
+   - Check for deprecated Express APIs
+   - Verify middleware usage is correct for Express 5
+   - Check request/response handling patterns
+
+4. **Type Safety**
+   - Verify TypeScript types are correctly used
+   - Check for `any` types that should be properly typed
+   - Verify type imports/exports are correct
+   - Check for missing type definitions
+
+5. **Build System Compatibility**
+   - Verify Vite-specific code (imports, config) is valid
+   - Check Electron-specific APIs are used correctly
+   - Verify module resolution paths are correct
+
+6. **Package Dependencies**
+   - Check for imports from packages not in package.json
+   - Verify version compatibility between dependencies
+   - Check for circular dependencies
+
+Provide a detailed report with:
+- File paths and line numbers of invalid code
+- Specific error description (what's wrong and why)
+- Expected vs actual behavior
+- Priority level (CRITICAL for build-breaking issues)
+```
+
+#### Agent 2: Security Vulnerability Scanner
+
+**Focus:** Security issues and vulnerabilities
+
+**Instructions for Agent 2:**
+
+```
+Analyze the git diff for security vulnerabilities:
+
+1. **Injection Vulnerabilities**
+   - SQL injection (if applicable)
+   - Command injection (exec, spawn, etc.)
+   - Path traversal vulnerabilities
+   - XSS vulnerabilities in React components
+
+2. **Authentication & Authorization**
+   - Missing authentication checks
+   - Insecure token handling
+   - Authorization bypasses
+   - Session management issues
+
+3. **Data Handling**
+   - Unsafe deserialization
+   - Insecure file operations
+   - Missing input validation
+   - Sensitive data exposure (secrets, tokens, passwords)
+
+4. **Dependencies**
+   - Known vulnerable packages
+   - Insecure dependency versions
+   - Missing security patches
+
+5. **API Security**
+   - Missing CORS configuration
+   - Insecure API endpoints
+   - Missing rate limiting
+   - Insecure WebSocket connections
+
+6. **Electron-Specific**
+   - Insecure IPC communication
+   - Missing context isolation checks
+   - Insecure preload scripts
+   - Missing CSP headers
+
+Provide a detailed report with:
+- Vulnerability type and severity (CRITICAL, HIGH, MEDIUM, LOW)
+- File paths and line numbers
+- Attack vector description
+- Recommended fix approach
+```
+
+#### Agent 3: Code Quality & Clean Code
+
+**Focus:** Dirty code, code smells, and quality issues
+
+**Instructions for Agent 3:**
+
+```
+Analyze the git diff for code quality issues:
+
+1. **Code Smells**
+   - Long functions/methods (>50 lines)
+   - High cyclomatic complexity
+   - Duplicate code
+   - Dead code
+   - Magic numbers/strings
+
+2. **Best Practices**
+   - Missing error handling
+   - Inconsistent naming conventions
+   - Poor separation of concerns
+   - Tight coupling
+   - Missing comments for complex logic
+
+3. **Performance Issues**
+   - Inefficient algorithms
+   - Memory leaks (event listeners, subscriptions)
+   - Unnecessary re-renders in React
+   - Missing memoization where needed
+   - Inefficient database queries (if applicable)
+
+4. **Maintainability**
+   - Hard-coded values
+   - Missing type definitions
+   - Inconsistent code style
+   - Poor file organization
+   - Missing tests for new code
+
+5. **React-Specific**
+   - Missing key props in lists
+   - Direct state mutations
+   - Missing cleanup in useEffect
+   - Unnecessary useState/useEffect
+   - Prop drilling issues
+
+Provide a detailed report with:
+- Issue type and severity
+- File paths and line numbers
+- Description of the problem
+- Impact on maintainability/performance
+- Recommended refactoring approach
+```
+
+#### Agent 4: Implementation Correctness
+
+**Focus:** Verify code implements requirements correctly
+
+**Instructions for Agent 4:**
+
+```
+Analyze the git diff for implementation correctness:
+
+1. **Logic Errors**
+   - Incorrect conditional logic
+   - Wrong variable usage
+   - Off-by-one errors
+   - Race conditions
+   - Missing null/undefined checks
+
+2. **Functional Requirements**
+   - Missing features from requirements
+   - Incorrect feature implementation
+   - Edge cases not handled
+   - Missing validation
+
+3. **Integration Issues**
+   - Incorrect API usage
+   - Wrong data format handling
+   - Missing error handling for external calls
+   - Incorrect state management
+
+4. **Type Errors**
+   - Type mismatches
+   - Missing type guards
+   - Incorrect type assertions
+   - Unsafe type operations
+
+5. **Testing Gaps**
+   - Missing unit tests
+   - Missing integration tests
+   - Tests don't cover edge cases
+   - Tests are incorrect
+
+Provide a detailed report with:
+- Issue description
+- File paths and line numbers
+- Expected vs actual behavior
+- Steps to reproduce (if applicable)
+- Recommended fix
+```
+
+#### Agent 5: Architecture & Design Patterns
+
+**Focus:** Architectural issues and design pattern violations
+
+**Instructions for Agent 5:**
+
+```
+Analyze the git diff for architectural and design issues:
+
+1. **Architecture Violations**
+   - Violation of project structure patterns
+   - Incorrect layer separation
+   - Missing abstractions
+   - Tight coupling between modules
+
+2. **Design Patterns**
+   - Incorrect pattern usage
+   - Missing patterns where needed
+   - Anti-patterns
+
+3. **Project-Specific Patterns**
+   - Check against project documentation (docs/ folder)
+   - Verify route organization (server routes)
+   - Check provider patterns (server providers)
+   - Verify component organization (UI components)
+
+4. **API Design**
+   - RESTful API violations
+   - Inconsistent response formats
+   - Missing error handling
+   - Incorrect status codes
+
+5. **State Management**
+   - Incorrect state management patterns
+   - Missing state normalization
+   - Inefficient state updates
+
+Provide a detailed report with:
+- Architectural issue description
+- File paths and affected areas
+- Impact on system design
+- Recommended architectural changes
+```
+
+### Phase 3: Consolidate Findings
+
+After all 5 deep dive agents complete their analysis:
+
+1. **Collect all findings** from each agent
+2. **Prioritize issues**:
+   - CRITICAL: Tech stack invalid code (build-breaking)
+   - HIGH: Security vulnerabilities, critical logic errors
+   - MEDIUM: Code quality issues, architectural problems
+   - LOW: Minor code smells, style issues
+
+3. **Group by file** to understand impact per file
+4. **Create a master report** summarizing all findings
+
+### Phase 4: Deepcode Fixes (5 Agents)
+
+Launch 5 deepcode agents to fix the issues found. Each agent should be invoked with the `@deepcode` agent.
+
+#### Deepcode Agent 1: Fix Tech Stack Invalid Code
+
+**Priority:** CRITICAL - Fix first
+
+**Instructions:**
+
+```
+Fix all invalid code based on tech stack issues identified by Agent 1.
+
+Focus on:
+1. Fixing TypeScript syntax errors
+2. Updating deprecated Node.js APIs
+3. Fixing React 19 compatibility issues
+4. Correcting Express 5 API usage
+5. Fixing type errors
+6. Resolving build-breaking issues
+
+After fixes, verify:
+- Code compiles without errors
+- TypeScript types are correct
+- No deprecated API usage
+```
+
+#### Deepcode Agent 2: Fix Security Vulnerabilities
+
+**Priority:** HIGH
+
+**Instructions:**
+
+```
+Fix all security vulnerabilities identified by Agent 2.
+
+Focus on:
+1. Adding input validation
+2. Fixing injection vulnerabilities
+3. Securing authentication/authorization
+4. Fixing insecure data handling
+5. Updating vulnerable dependencies
+6. Securing Electron IPC
+
+After fixes, verify:
+- Security vulnerabilities are addressed
+- No sensitive data exposure
+- Proper authentication/authorization
+```
+
+#### Deepcode Agent 3: Refactor Dirty Code
+
+**Priority:** MEDIUM
+
+**Instructions:**
+
+```
+Refactor code quality issues identified by Agent 3.
+
+Focus on:
+1. Extracting long functions
+2. Reducing complexity
+3. Removing duplicate code
+4. Adding error handling
+5. Improving React component structure
+6. Adding missing comments
+
+After fixes, verify:
+- Code follows best practices
+- No code smells remain
+- Performance optimizations applied
+```
+
+#### Deepcode Agent 4: Fix Implementation Errors
+
+**Priority:** HIGH
+
+**Instructions:**
+
+```
+Fix implementation correctness issues identified by Agent 4.
+
+Focus on:
+1. Fixing logic errors
+2. Adding missing features
+3. Handling edge cases
+4. Fixing type errors
+5. Adding missing tests
+
+After fixes, verify:
+- Logic is correct
+- Edge cases handled
+- Tests pass
+```
+
+#### Deepcode Agent 5: Fix Architectural Issues
+
+**Priority:** MEDIUM
+
+**Instructions:**
+
+```
+Fix architectural issues identified by Agent 5.
+
+Focus on:
+1. Correcting architecture violations
+2. Applying proper design patterns
+3. Fixing API design issues
+4. Improving state management
+5. Following project patterns
+
+After fixes, verify:
+- Architecture is sound
+- Patterns are correctly applied
+- Code follows project structure
+```
+
+### Phase 5: Verification
+
+After all fixes are complete:
+
+1. **Run TypeScript compilation check**
+
+   ```bash
+   npm run build:packages
+   ```
+
+2. **Run linting**
+
+   ```bash
+   npm run lint
+   ```
+
+3. **Run tests** (if applicable)
+
+   ```bash
+   npm run test:server
+   npm run test
+   ```
+
+4. **Verify git diff** shows only intended changes
+
+   ```bash
+   git diff HEAD
+   ```
+
+5. **Create summary report**:
+   - Issues found by each agent
+   - Issues fixed by each agent
+   - Remaining issues (if any)
+   - Verification results
+
+## Workflow Summary
+
+1. ✅ Accept optional target branch argument (defaults to main/master if not provided)
+2. ✅ Determine current branch and target branch (from argument or auto-detect main/master)
+3. ✅ Get git diff comparing current branch against target branch (`git diff $TARGET_REF...HEAD`)
+4. ✅ Include uncommitted changes in analysis (`git diff HEAD`, `git diff --cached`)
+5. ✅ Launch 5 deep dive agents (parallel analysis) with branch diff
+6. ✅ Consolidate findings and prioritize
+7. ✅ Launch 5 deepcode agents (sequential fixes, priority order)
+8. ✅ Verify fixes with build/lint/test
+9. ✅ Report summary
+
+## Notes
+
+- **Tech stack validation is HIGHEST PRIORITY** - invalid code must be fixed first
+- **Target branch argument**: The command accepts an optional target branch name as the first argument. If not provided, it automatically detects and uses `main` or `master` (in that order)
+- Each deep dive agent should work independently and provide comprehensive analysis
+- Deepcode agents should fix issues in priority order
+- All fixes should maintain existing functionality
+- If an agent finds no issues in their domain, they should report "No issues found"
+- If fixes introduce new issues, they should be caught in verification phase
+- The target branch is validated to ensure it exists (locally or remotely) before proceeding with the review

.claude/commands/review.md

@@ -0,0 +1,484 @@
+# Code Review Command
+
+Comprehensive code review using multiple deep dive agents to analyze git diff for correctness, security, code quality, and tech stack compliance, followed by automated fixes using deepcode agents.
+
+## Usage
+
+This command analyzes all changes in the git diff and verifies:
+
+1. **Invalid code based on tech stack** (HIGHEST PRIORITY)
+2. Security vulnerabilities
+3. Code quality issues (dirty code)
+4. Implementation correctness
+
+Then automatically fixes any issues found.
+
+## Instructions
+
+### Phase 1: Get Git Diff
+
+1. **Get the current git diff**
+
+   ```bash
+   git diff HEAD
+   ```
+
+   If you need staged changes instead:
+
+   ```bash
+   git diff --cached
+   ```
+
+   Or for a specific commit range:
+
+   ```bash
+   git diff <base-branch>
+   ```
+
+2. **Get list of changed files**
+
+   ```bash
+   git diff --name-only HEAD
+   ```
+
+3. **Understand the tech stack** (for validation):
+   - **Node.js**: >=22.0.0 <23.0.0
+   - **TypeScript**: 5.9.3
+   - **React**: 19.2.3
+   - **Express**: 5.2.1
+   - **Electron**: 39.2.7
+   - **Vite**: 7.3.0
+   - **Vitest**: 4.0.16
+   - Check `package.json` files for exact versions
+
+### Phase 2: Deep Dive Analysis (5 Agents)
+
+Launch 5 separate deep dive agents, each with a specific focus area. Each agent should be invoked with the `@deepdive` agent and given the git diff along with their specific instructions.
+
+#### Agent 1: Tech Stack Validation (HIGHEST PRIORITY)
+
+**Focus:** Verify code is valid for the tech stack
+
+**Instructions for Agent 1:**
+
+```
+Analyze the git diff for invalid code based on the tech stack:
+
+1. **TypeScript/JavaScript Syntax**
+   - Check for valid TypeScript syntax (no invalid type annotations, correct import/export syntax)
+   - Verify Node.js API usage is compatible with Node.js >=22.0.0 <23.0.0
+   - Check for deprecated APIs or features not available in the Node.js version
+   - Verify ES module syntax (type: "module" in package.json)
+
+2. **React 19.2.3 Compatibility**
+   - Check for deprecated React APIs or patterns
+   - Verify hooks usage is correct for React 19
+   - Check for invalid JSX syntax
+   - Verify component patterns match React 19 conventions
+
+3. **Express 5.2.1 Compatibility**
+   - Check for deprecated Express APIs
+   - Verify middleware usage is correct for Express 5
+   - Check request/response handling patterns
+
+4. **Type Safety**
+   - Verify TypeScript types are correctly used
+   - Check for `any` types that should be properly typed
+   - Verify type imports/exports are correct
+   - Check for missing type definitions
+
+5. **Build System Compatibility**
+   - Verify Vite-specific code (imports, config) is valid
+   - Check Electron-specific APIs are used correctly
+   - Verify module resolution paths are correct
+
+6. **Package Dependencies**
+   - Check for imports from packages not in package.json
+   - Verify version compatibility between dependencies
+   - Check for circular dependencies
+
+Provide a detailed report with:
+- File paths and line numbers of invalid code
+- Specific error description (what's wrong and why)
+- Expected vs actual behavior
+- Priority level (CRITICAL for build-breaking issues)
+```
+
+#### Agent 2: Security Vulnerability Scanner
+
+**Focus:** Security issues and vulnerabilities
+
+**Instructions for Agent 2:**
+
+```
+Analyze the git diff for security vulnerabilities:
+
+1. **Injection Vulnerabilities**
+   - SQL injection (if applicable)
+   - Command injection (exec, spawn, etc.)
+   - Path traversal vulnerabilities
+   - XSS vulnerabilities in React components
+
+2. **Authentication & Authorization**
+   - Missing authentication checks
+   - Insecure token handling
+   - Authorization bypasses
+   - Session management issues
+
+3. **Data Handling**
+   - Unsafe deserialization
+   - Insecure file operations
+   - Missing input validation
+   - Sensitive data exposure (secrets, tokens, passwords)
+
+4. **Dependencies**
+   - Known vulnerable packages
+   - Insecure dependency versions
+   - Missing security patches
+
+5. **API Security**
+   - Missing CORS configuration
+   - Insecure API endpoints
+   - Missing rate limiting
+   - Insecure WebSocket connections
+
+6. **Electron-Specific**
+   - Insecure IPC communication
+   - Missing context isolation checks
+   - Insecure preload scripts
+   - Missing CSP headers
+
+Provide a detailed report with:
+- Vulnerability type and severity (CRITICAL, HIGH, MEDIUM, LOW)
+- File paths and line numbers
+- Attack vector description
+- Recommended fix approach
+```
+
+#### Agent 3: Code Quality & Clean Code
+
+**Focus:** Dirty code, code smells, and quality issues
+
+**Instructions for Agent 3:**
+
+```
+Analyze the git diff for code quality issues:
+
+1. **Code Smells**
+   - Long functions/methods (>50 lines)
+   - High cyclomatic complexity
+   - Duplicate code
+   - Dead code
+   - Magic numbers/strings
+
+2. **Best Practices**
+   - Missing error handling
+   - Inconsistent naming conventions
+   - Poor separation of concerns
+   - Tight coupling
+   - Missing comments for complex logic
+
+3. **Performance Issues**
+   - Inefficient algorithms
+   - Memory leaks (event listeners, subscriptions)
+   - Unnecessary re-renders in React
+   - Missing memoization where needed
+   - Inefficient database queries (if applicable)
+
+4. **Maintainability**
+   - Hard-coded values
+   - Missing type definitions
+   - Inconsistent code style
+   - Poor file organization
+   - Missing tests for new code
+
+5. **React-Specific**
+   - Missing key props in lists
+   - Direct state mutations
+   - Missing cleanup in useEffect
+   - Unnecessary useState/useEffect
+   - Prop drilling issues
+
+Provide a detailed report with:
+- Issue type and severity
+- File paths and line numbers
+- Description of the problem
+- Impact on maintainability/performance
+- Recommended refactoring approach
+```
+
+#### Agent 4: Implementation Correctness
+
+**Focus:** Verify code implements requirements correctly
+
+**Instructions for Agent 4:**
+
+```
+Analyze the git diff for implementation correctness:
+
+1. **Logic Errors**
+   - Incorrect conditional logic
+   - Wrong variable usage
+   - Off-by-one errors
+   - Race conditions
+   - Missing null/undefined checks
+
+2. **Functional Requirements**
+   - Missing features from requirements
+   - Incorrect feature implementation
+   - Edge cases not handled
+   - Missing validation
+
+3. **Integration Issues**
+   - Incorrect API usage
+   - Wrong data format handling
+   - Missing error handling for external calls
+   - Incorrect state management
+
+4. **Type Errors**
+   - Type mismatches
+   - Missing type guards
+   - Incorrect type assertions
+   - Unsafe type operations
+
+5. **Testing Gaps**
+   - Missing unit tests
+   - Missing integration tests
+   - Tests don't cover edge cases
+   - Tests are incorrect
+
+Provide a detailed report with:
+- Issue description
+- File paths and line numbers
+- Expected vs actual behavior
+- Steps to reproduce (if applicable)
+- Recommended fix
+```
+
+#### Agent 5: Architecture & Design Patterns
+
+**Focus:** Architectural issues and design pattern violations
+
+**Instructions for Agent 5:**
+
+```
+Analyze the git diff for architectural and design issues:
+
+1. **Architecture Violations**
+   - Violation of project structure patterns
+   - Incorrect layer separation
+   - Missing abstractions
+   - Tight coupling between modules
+
+2. **Design Patterns**
+   - Incorrect pattern usage
+   - Missing patterns where needed
+   - Anti-patterns
+
+3. **Project-Specific Patterns**
+   - Check against project documentation (docs/ folder)
+   - Verify route organization (server routes)
+   - Check provider patterns (server providers)
+   - Verify component organization (UI components)
+
+4. **API Design**
+   - RESTful API violations
+   - Inconsistent response formats
+   - Missing error handling
+   - Incorrect status codes
+
+5. **State Management**
+   - Incorrect state management patterns
+   - Missing state normalization
+   - Inefficient state updates
+
+Provide a detailed report with:
+- Architectural issue description
+- File paths and affected areas
+- Impact on system design
+- Recommended architectural changes
+```
+
+### Phase 3: Consolidate Findings
+
+After all 5 deep dive agents complete their analysis:
+
+1. **Collect all findings** from each agent
+2. **Prioritize issues**:
+   - CRITICAL: Tech stack invalid code (build-breaking)
+   - HIGH: Security vulnerabilities, critical logic errors
+   - MEDIUM: Code quality issues, architectural problems
+   - LOW: Minor code smells, style issues
+
+3. **Group by file** to understand impact per file
+4. **Create a master report** summarizing all findings
+
+### Phase 4: Deepcode Fixes (5 Agents)
+
+Launch 5 deepcode agents to fix the issues found. Each agent should be invoked with the `@deepcode` agent.
+
+#### Deepcode Agent 1: Fix Tech Stack Invalid Code
+
+**Priority:** CRITICAL - Fix first
+
+**Instructions:**
+
+```
+Fix all invalid code based on tech stack issues identified by Agent 1.
+
+Focus on:
+1. Fixing TypeScript syntax errors
+2. Updating deprecated Node.js APIs
+3. Fixing React 19 compatibility issues
+4. Correcting Express 5 API usage
+5. Fixing type errors
+6. Resolving build-breaking issues
+
+After fixes, verify:
+- Code compiles without errors
+- TypeScript types are correct
+- No deprecated API usage
+```
+
+#### Deepcode Agent 2: Fix Security Vulnerabilities
+
+**Priority:** HIGH
+
+**Instructions:**
+
+```
+Fix all security vulnerabilities identified by Agent 2.
+
+Focus on:
+1. Adding input validation
+2. Fixing injection vulnerabilities
+3. Securing authentication/authorization
+4. Fixing insecure data handling
+5. Updating vulnerable dependencies
+6. Securing Electron IPC
+
+After fixes, verify:
+- Security vulnerabilities are addressed
+- No sensitive data exposure
+- Proper authentication/authorization
+```
+
+#### Deepcode Agent 3: Refactor Dirty Code
+
+**Priority:** MEDIUM
+
+**Instructions:**
+
+```
+Refactor code quality issues identified by Agent 3.
+
+Focus on:
+1. Extracting long functions
+2. Reducing complexity
+3. Removing duplicate code
+4. Adding error handling
+5. Improving React component structure
+6. Adding missing comments
+
+After fixes, verify:
+- Code follows best practices
+- No code smells remain
+- Performance optimizations applied
+```
+
+#### Deepcode Agent 4: Fix Implementation Errors
+
+**Priority:** HIGH
+
+**Instructions:**
+
+```
+Fix implementation correctness issues identified by Agent 4.
+
+Focus on:
+1. Fixing logic errors
+2. Adding missing features
+3. Handling edge cases
+4. Fixing type errors
+5. Adding missing tests
+
+After fixes, verify:
+- Logic is correct
+- Edge cases handled
+- Tests pass
+```
+
+#### Deepcode Agent 5: Fix Architectural Issues
+
+**Priority:** MEDIUM
+
+**Instructions:**
+
+```
+Fix architectural issues identified by Agent 5.
+
+Focus on:
+1. Correcting architecture violations
+2. Applying proper design patterns
+3. Fixing API design issues
+4. Improving state management
+5. Following project patterns
+
+After fixes, verify:
+- Architecture is sound
+- Patterns are correctly applied
+- Code follows project structure
+```
+
+### Phase 5: Verification
+
+After all fixes are complete:
+
+1. **Run TypeScript compilation check**
+
+   ```bash
+   npm run build:packages
+   ```
+
+2. **Run linting**
+
+   ```bash
+   npm run lint
+   ```
+
+3. **Run tests** (if applicable)
+
+   ```bash
+   npm run test:server
+   npm run test
+   ```
+
+4. **Verify git diff** shows only intended changes
+
+   ```bash
+   git diff HEAD
+   ```
+
+5. **Create summary report**:
+   - Issues found by each agent
+   - Issues fixed by each agent
+   - Remaining issues (if any)
+   - Verification results
+
+## Workflow Summary
+
+1. ✅ Get git diff
+2. ✅ Launch 5 deep dive agents (parallel analysis)
+3. ✅ Consolidate findings and prioritize
+4. ✅ Launch 5 deepcode agents (sequential fixes, priority order)
+5. ✅ Verify fixes with build/lint/test
+6. ✅ Report summary
+
+## Notes
+
+- **Tech stack validation is HIGHEST PRIORITY** - invalid code must be fixed first
+- Each deep dive agent should work independently and provide comprehensive analysis
+- Deepcode agents should fix issues in priority order
+- All fixes should maintain existing functionality
+- If an agent finds no issues in their domain, they should report "No issues found"
+- If fixes introduce new issues, they should be caught in verification phase

.claude/commands/thorough.md

@@ -0,0 +1,45 @@
+When you think you are done, you are NOT done.
+
+You must run a mandatory 3-pass verification before concluding:
+
+## Pass 1: Correctness & Functionality
+
+- [ ] Verify logic matches requirements and specifications
+- [ ] Check type safety (TypeScript types are correct and complete)
+- [ ] Ensure imports are correct and follow project conventions
+- [ ] Verify all functions/classes work as intended
+- [ ] Check that return values and side effects are correct
+- [ ] Run relevant tests if they exist, or verify testability
+- [ ] Confirm integration with existing code works properly
+
+## Pass 2: Edge Cases & Safety
+
+- [ ] Handle null/undefined inputs gracefully
+- [ ] Validate all user inputs and external data
+- [ ] Check error handling (try/catch, error boundaries, etc.)
+- [ ] Verify security considerations (no sensitive data exposure, proper auth checks)
+- [ ] Test boundary conditions (empty arrays, zero values, max lengths, etc.)
+- [ ] Ensure resource cleanup (file handles, connections, timers)
+- [ ] Check for potential race conditions or async issues
+- [ ] Verify file path security (no directory traversal vulnerabilities)
+
+## Pass 3: Maintainability & Code Quality
+
+- [ ] Code follows project style guide and conventions
+- [ ] Functions/classes are single-purpose and well-named
+- [ ] Remove dead code, unused imports, and console.logs
+- [ ] Extract magic numbers/strings into named constants
+- [ ] Check for code duplication (DRY principle)
+- [ ] Verify appropriate abstraction levels (not over/under-engineered)
+- [ ] Add necessary comments for complex logic
+- [ ] Ensure consistent error messages and logging
+- [ ] Check that code is readable and self-documenting
+- [ ] Verify proper separation of concerns
+
+**For each pass, explicitly report:**
+
+- What you checked
+- Any issues found and how they were fixed
+- Any remaining concerns or trade-offs
+
+Only after completing all three passes with explicit findings may you conclude the work is done.

.dockerignore

@@ -0,0 +1,19 @@
+# Dependencies
+node_modules/
+**/node_modules/
+
+# Build outputs
+dist/
+**/dist/
+dist-electron/
+**/dist-electron/
+build/
+**/build/
+.next/
+**/.next/
+.nuxt/
+**/.nuxt/
+out/
+**/out/
+.cache/
+**/.cache/

.github/workflows/e2e-tests.yml

@@ -31,24 +31,99 @@ jobs:
       - name: Build server
         run: npm run build --workspace=apps/server
 
+      - name: Set up Git user
+        run: |
+          git config --global user.name "GitHub CI"
+          git config --global user.email "ci@example.com"
+
       - name: Start backend server
-        run: npm run start --workspace=apps/server &
+        run: |
+          echo "Starting backend server..."
+          # Start server in background and save PID
+          npm run start --workspace=apps/server > backend.log 2>&1 &
+          SERVER_PID=$!
+          echo "Server started with PID: $SERVER_PID"
+          echo "SERVER_PID=$SERVER_PID" >> $GITHUB_ENV
+
         env:
           PORT: 3008
           NODE_ENV: test
+          # Use a deterministic API key so Playwright can log in reliably
+          AUTOMAKER_API_KEY: test-api-key-for-e2e-tests
+          # Reduce log noise in CI
+          AUTOMAKER_HIDE_API_KEY: 'true'
+          # Avoid real API calls during CI
+          AUTOMAKER_MOCK_AGENT: 'true'
+          # Simulate containerized environment to skip sandbox confirmation dialogs
+          IS_CONTAINERIZED: 'true'
 
       - name: Wait for backend server
         run: |
           echo "Waiting for backend server to be ready..."
-          for i in {1..30}; do
-            if curl -s http://localhost:3008/api/health > /dev/null 2>&1; then
+
+          # Check if server process is running
+          if [ -z "$SERVER_PID" ]; then
+            echo "ERROR: Server PID not found in environment"
+            cat backend.log 2>/dev/null || echo "No backend log found"
+            exit 1
+          fi
+
+          # Check if process is actually running
+          if ! kill -0 $SERVER_PID 2>/dev/null; then
+            echo "ERROR: Server process $SERVER_PID is not running!"
+            echo "=== Backend logs ==="
+            cat backend.log
+            echo ""
+            echo "=== Recent system logs ==="
+            dmesg 2>/dev/null | tail -20 || echo "No dmesg available"
+            exit 1
+          fi
+
+          # Wait for health endpoint
+          for i in {1..60}; do
+            if curl -s -f http://localhost:3008/api/health > /dev/null 2>&1; then
               echo "Backend server is ready!"
+              echo "=== Backend logs ==="
+              cat backend.log
+              echo ""
+              echo "Health check response:"
+              curl -s http://localhost:3008/api/health | jq . 2>/dev/null || echo "Health check: $(curl -s http://localhost:3008/api/health 2>/dev/null || echo 'No response')"
               exit 0
             fi
-            echo "Waiting... ($i/30)"
+            
+            # Check if server process is still running
+            if ! kill -0 $SERVER_PID 2>/dev/null; then
+              echo "ERROR: Server process died during wait!"
+              echo "=== Backend logs ==="
+              cat backend.log
+              exit 1
+            fi
+            
+            echo "Waiting... ($i/60)"
             sleep 1
           done
-          echo "Backend server failed to start!"
+
+          echo "ERROR: Backend server failed to start within 60 seconds!"
+          echo "=== Backend logs ==="
+          cat backend.log
+          echo ""
+          echo "=== Process status ==="
+          ps aux | grep -E "(node|tsx)" | grep -v grep || echo "No node processes found"
+          echo ""
+          echo "=== Port status ==="
+          netstat -tlnp 2>/dev/null | grep :3008 || echo "Port 3008 not listening"
+          lsof -i :3008 2>/dev/null || echo "lsof not available or port not in use"
+          echo ""
+          echo "=== Health endpoint test ==="
+          curl -v http://localhost:3008/api/health 2>&1 || echo "Health endpoint failed"
+
+          # Kill the server process if it's still hanging
+          if kill -0 $SERVER_PID 2>/dev/null; then
+            echo ""
+            echo "Killing stuck server process..."
+            kill -9 $SERVER_PID 2>/dev/null || true
+          fi
+
           exit 1
 
       - name: Run E2E tests
@@ -59,6 +134,20 @@ jobs:
           CI: true
           VITE_SERVER_URL: http://localhost:3008
           VITE_SKIP_SETUP: 'true'
+          # Keep UI-side login/defaults consistent
+          AUTOMAKER_API_KEY: test-api-key-for-e2e-tests
+
+      - name: Print backend logs on failure
+        if: failure()
+        run: |
+          echo "=== E2E Tests Failed - Backend Logs ==="
+          cat backend.log 2>/dev/null || echo "No backend log found"
+          echo ""
+          echo "=== Process status at failure ==="
+          ps aux | grep -E "(node|tsx)" | grep -v grep || echo "No node processes found"
+          echo ""
+          echo "=== Port status ==="
+          netstat -tlnp 2>/dev/null | grep :3008 || echo "Port 3008 not listening"
 
       - name: Upload Playwright report
         uses: actions/upload-artifact@v4
@@ -68,10 +157,22 @@ jobs:
           path: apps/ui/playwright-report/
           retention-days: 7
 
-      - name: Upload test results
+      - name: Upload test results (screenshots, traces, videos)
         uses: actions/upload-artifact@v4
-        if: failure()
+        if: always()
         with:
           name: test-results
-          path: apps/ui/test-results/
+          path: |
+            apps/ui/test-results/
           retention-days: 7
+          if-no-files-found: ignore
+
+      - name: Cleanup - Kill backend server
+        if: always()
+        run: |
+          if [ -n "$SERVER_PID" ]; then
+            echo "Cleaning up backend server (PID: $SERVER_PID)..."
+            kill $SERVER_PID 2>/dev/null || true
+            kill -9 $SERVER_PID 2>/dev/null || true
+            echo "Backend server cleanup complete"
+          fi

.github/workflows/security-audit.yml

@@ -26,5 +26,5 @@ jobs:
           check-lockfile: 'true'
 
       - name: Run npm audit
-        run: npm audit --audit-level=moderate
+        run: npm audit --audit-level=critical
         continue-on-error: false

.gitignore

@@ -73,6 +73,9 @@ blob-report/
 !.env.example
 !.env.local.example
 
+# Codex config (contains API keys)
+.codex/config.toml
+
 # TypeScript
 *.tsbuildinfo
 
@@ -81,6 +84,15 @@ blob-report/
 
 docker-compose.override.yml
 .claude/docker-compose.override.yml
+.claude/hans/
 
 pnpm-lock.yaml
 yarn.lock
+
+# Fork-specific workflow files (should never be committed)
+DEVELOPMENT_WORKFLOW.md
+check-sync.sh
+# API key files
+data/.api-key
+data/credentials.json
+data/

.husky/pre-commit

@@ -1 +1,46 @@
-npx lint-staged
+#!/usr/bin/env sh
+
+# Try to load nvm if available (optional - works without it too)
+if [ -z "$NVM_DIR" ]; then
+  # Check for Herd's nvm first (macOS with Herd)
+  if [ -s "$HOME/Library/Application Support/Herd/config/nvm/nvm.sh" ]; then
+    export NVM_DIR="$HOME/Library/Application Support/Herd/config/nvm"
+  # Then check standard nvm location
+  elif [ -s "$HOME/.nvm/nvm.sh" ]; then
+    export NVM_DIR="$HOME/.nvm"
+  fi
+fi
+
+# Source nvm if found (silently skip if not available)
+[ -n "$NVM_DIR" ] && [ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh" 2>/dev/null
+
+# Load node version from .nvmrc if using nvm (silently skip if nvm not available or fails)
+if [ -f .nvmrc ] && command -v nvm >/dev/null 2>&1; then
+  # Check if Unix nvm was sourced (it's a shell function with NVM_DIR set)
+  if [ -n "$NVM_DIR" ] && type nvm 2>/dev/null | grep -q "function"; then
+    # Unix nvm: reads .nvmrc automatically
+    nvm use >/dev/null 2>&1 || true
+  else
+    # nvm-windows: needs explicit version from .nvmrc
+    NODE_VERSION=$(cat .nvmrc | tr -d '[:space:]')
+    if [ -n "$NODE_VERSION" ]; then
+      nvm use "$NODE_VERSION" >/dev/null 2>&1 || true
+    fi
+  fi
+fi
+
+# Ensure common system paths are in PATH (for systems without nvm)
+# This helps find node/npm installed via Homebrew, system packages, etc.
+export PATH="$PATH:/usr/local/bin:/opt/homebrew/bin:/usr/bin"
+
+# Run lint-staged - works with or without nvm
+# Prefer npx, fallback to npm exec, both work with system-installed Node.js
+if command -v npx >/dev/null 2>&1; then
+  npx lint-staged
+elif command -v npm >/dev/null 2>&1; then
+  npm exec -- lint-staged
+else
+  echo "Error: Neither npx nor npm found in PATH."
+  echo "Please ensure Node.js is installed (via nvm, Homebrew, system package manager, etc.)"
+  exit 1
+fi

.prettierignore

@@ -23,6 +23,8 @@ pnpm-lock.yaml
 # Generated files
 *.min.js
 *.min.css
+routeTree.gen.ts
+apps/ui/src/routeTree.gen.ts
 
 # Test artifacts
 test-results/

Dockerfile

@@ -8,10 +8,12 @@
 # =============================================================================
 # BASE STAGE - Common setup for all builds (DRY: defined once, used by all)
 # =============================================================================
-FROM node:22-alpine AS base
+FROM node:22-slim AS base
 
 # Install build dependencies for native modules (node-pty)
-RUN apk add --no-cache python3 make g++
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    python3 make g++ \
+    && rm -rf /var/lib/apt/lists/*
 
 WORKDIR /app
 
@@ -51,30 +53,63 @@ RUN npm run build:packages && npm run build --workspace=apps/server
 # =============================================================================
 # SERVER PRODUCTION STAGE
 # =============================================================================
-FROM node:22-alpine AS server
+FROM node:22-slim AS server
 
-# Install git, curl, bash (for terminal), and GitHub CLI (pinned version, multi-arch)
-RUN apk add --no-cache git curl bash && \
-    GH_VERSION="2.63.2" && \
-    ARCH=$(uname -m) && \
-    case "$ARCH" in \
+# Build argument for tracking which commit this image was built from
+ARG GIT_COMMIT_SHA=unknown
+LABEL automaker.git.commit.sha="${GIT_COMMIT_SHA}"
+
+# Install git, curl, bash (for terminal), gosu (for user switching), and GitHub CLI (pinned version, multi-arch)
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    git curl bash gosu ca-certificates openssh-client \
+    && GH_VERSION="2.63.2" \
+    && ARCH=$(uname -m) \
+    && case "$ARCH" in \
         x86_64) GH_ARCH="amd64" ;; \
         aarch64|arm64) GH_ARCH="arm64" ;; \
         *) echo "Unsupported architecture: $ARCH" && exit 1 ;; \
-    esac && \
-    curl -L "https://github.com/cli/cli/releases/download/v${GH_VERSION}/gh_${GH_VERSION}_linux_${GH_ARCH}.tar.gz" -o gh.tar.gz && \
-    tar -xzf gh.tar.gz && \
-    mv gh_${GH_VERSION}_linux_${GH_ARCH}/bin/gh /usr/local/bin/gh && \
-    rm -rf gh.tar.gz gh_${GH_VERSION}_linux_${GH_ARCH}
+    esac \
+    && curl -L "https://github.com/cli/cli/releases/download/v${GH_VERSION}/gh_${GH_VERSION}_linux_${GH_ARCH}.tar.gz" -o gh.tar.gz \
+    && tar -xzf gh.tar.gz \
+    && mv gh_${GH_VERSION}_linux_${GH_ARCH}/bin/gh /usr/local/bin/gh \
+    && rm -rf gh.tar.gz gh_${GH_VERSION}_linux_${GH_ARCH} \
+    && rm -rf /var/lib/apt/lists/*
 
-# Install Claude CLI globally
+# Install Claude CLI globally (available to all users via npm global bin)
 RUN npm install -g @anthropic-ai/claude-code
 
-WORKDIR /app
+# Create non-root user with home directory BEFORE installing Cursor CLI
+RUN groupadd -g 1001 automaker && \
+    useradd -u 1001 -g automaker -m -d /home/automaker -s /bin/bash automaker && \
+    mkdir -p /home/automaker/.local/bin && \
+    mkdir -p /home/automaker/.cursor && \
+    chown -R automaker:automaker /home/automaker && \
+    chmod 700 /home/automaker/.cursor
 
-# Create non-root user
-RUN addgroup -g 1001 -S automaker && \
-    adduser -S automaker -u 1001
+# Install Cursor CLI as the automaker user
+# Set HOME explicitly and install to /home/automaker/.local/bin/
+USER automaker
+ENV HOME=/home/automaker
+RUN curl https://cursor.com/install -fsS | bash && \
+    echo "=== Checking Cursor CLI installation ===" && \
+    ls -la /home/automaker/.local/bin/ && \
+    echo "=== PATH is: $PATH ===" && \
+    (which cursor-agent && cursor-agent --version) || echo "cursor-agent installed (may need auth setup)"
+USER root
+
+# Add PATH to profile so it's available in all interactive shells (for login shells)
+RUN mkdir -p /etc/profile.d && \
+    echo 'export PATH="/home/automaker/.local/bin:$PATH"' > /etc/profile.d/cursor-cli.sh && \
+    chmod +x /etc/profile.d/cursor-cli.sh
+
+# Add to automaker's .bashrc for bash interactive shells
+RUN echo 'export PATH="/home/automaker/.local/bin:$PATH"' >> /home/automaker/.bashrc && \
+    chown automaker:automaker /home/automaker/.bashrc
+
+# Also add to root's .bashrc since docker exec defaults to root
+RUN echo 'export PATH="/home/automaker/.local/bin:$PATH"' >> /root/.bashrc
+
+WORKDIR /app
 
 # Copy root package.json (needed for workspace resolution)
 COPY --from=server-builder /app/package*.json ./
@@ -98,12 +133,19 @@ RUN git config --system --add safe.directory '*' && \
     # Use gh as credential helper (works with GH_TOKEN env var)
     git config --system credential.helper '!gh auth git-credential'
 
-# Switch to non-root user
-USER automaker
+# Copy entrypoint script for fixing permissions on mounted volumes
+COPY docker-entrypoint.sh /usr/local/bin/docker-entrypoint.sh
+RUN chmod +x /usr/local/bin/docker-entrypoint.sh
+
+# Note: We stay as root here so entrypoint can fix permissions
+# The entrypoint script will switch to automaker user before running the command
 
 # Environment variables
 ENV PORT=3008
 ENV DATA_DIR=/data
+ENV HOME=/home/automaker
+# Add user's local bin to PATH for cursor-agent
+ENV PATH="/home/automaker/.local/bin:${PATH}"
 
 # Expose port
 EXPOSE 3008
@@ -112,6 +154,9 @@ EXPOSE 3008
 HEALTHCHECK --interval=30s --timeout=3s --start-period=5s --retries=3 \
     CMD curl -f http://localhost:3008/api/health || exit 1
 
+# Use entrypoint to fix permissions before starting
+ENTRYPOINT ["/usr/local/bin/docker-entrypoint.sh"]
+
 # Start server
 CMD ["node", "apps/server/dist/index.js"]
 
@@ -143,6 +188,10 @@ RUN npm run build:packages && npm run build --workspace=apps/ui
 # =============================================================================
 FROM nginx:alpine AS ui
 
+# Build argument for tracking which commit this image was built from
+ARG GIT_COMMIT_SHA=unknown
+LABEL automaker.git.commit.sha="${GIT_COMMIT_SHA}"
+
 # Copy built files
 COPY --from=ui-builder /app/apps/ui/dist /usr/share/nginx/html
 

Dockerfile.dev

@@ -0,0 +1,80 @@
+# Automaker Development Dockerfile
+# For development with live reload via volume mounting
+# Source code is NOT copied - it's mounted as a volume
+#
+# Usage:
+#   docker compose -f docker-compose.dev.yml up
+
+FROM node:22-slim
+
+# Install build dependencies for native modules (node-pty) and runtime tools
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    python3 make g++ \
+    git curl bash gosu ca-certificates openssh-client \
+    && GH_VERSION="2.63.2" \
+    && ARCH=$(uname -m) \
+    && case "$ARCH" in \
+        x86_64) GH_ARCH="amd64" ;; \
+        aarch64|arm64) GH_ARCH="arm64" ;; \
+        *) echo "Unsupported architecture: $ARCH" && exit 1 ;; \
+    esac \
+    && curl -L "https://github.com/cli/cli/releases/download/v${GH_VERSION}/gh_${GH_VERSION}_linux_${GH_ARCH}.tar.gz" -o gh.tar.gz \
+    && tar -xzf gh.tar.gz \
+    && mv gh_${GH_VERSION}_linux_${GH_ARCH}/bin/gh /usr/local/bin/gh \
+    && rm -rf gh.tar.gz gh_${GH_VERSION}_linux_${GH_ARCH} \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install Claude CLI globally
+RUN npm install -g @anthropic-ai/claude-code
+
+# Create non-root user
+RUN groupadd -g 1001 automaker && \
+    useradd -u 1001 -g automaker -m -d /home/automaker -s /bin/bash automaker && \
+    mkdir -p /home/automaker/.local/bin && \
+    mkdir -p /home/automaker/.cursor && \
+    chown -R automaker:automaker /home/automaker && \
+    chmod 700 /home/automaker/.cursor
+
+# Install Cursor CLI as automaker user
+USER automaker
+ENV HOME=/home/automaker
+RUN curl https://cursor.com/install -fsS | bash || true
+USER root
+
+# Add PATH to profile for Cursor CLI
+RUN mkdir -p /etc/profile.d && \
+    echo 'export PATH="/home/automaker/.local/bin:$PATH"' > /etc/profile.d/cursor-cli.sh && \
+    chmod +x /etc/profile.d/cursor-cli.sh
+
+# Add to user bashrc files
+RUN echo 'export PATH="/home/automaker/.local/bin:$PATH"' >> /home/automaker/.bashrc && \
+    chown automaker:automaker /home/automaker/.bashrc
+RUN echo 'export PATH="/home/automaker/.local/bin:$PATH"' >> /root/.bashrc
+
+WORKDIR /app
+
+# Create directories with proper permissions
+RUN mkdir -p /data /projects && chown automaker:automaker /data /projects
+
+# Configure git for mounted volumes
+RUN git config --system --add safe.directory '*' && \
+    git config --system credential.helper '!gh auth git-credential'
+
+# Copy entrypoint script
+COPY docker-entrypoint.sh /usr/local/bin/docker-entrypoint.sh
+RUN chmod +x /usr/local/bin/docker-entrypoint.sh
+
+# Environment variables
+ENV PORT=3008
+ENV DATA_DIR=/data
+ENV HOME=/home/automaker
+ENV PATH="/home/automaker/.local/bin:${PATH}"
+
+# Expose both dev ports
+EXPOSE 3007 3008
+
+# Use entrypoint for permission handling
+ENTRYPOINT ["/usr/local/bin/docker-entrypoint.sh"]
+
+# Default command - will be overridden by docker-compose
+CMD ["npm", "run", "dev:web"]

README.md

@@ -117,32 +117,32 @@ cd automaker
 # 2. Install dependencies
 npm install
 
-# 3. Build shared packages (Now can be skipped npm install / run dev does it automaticly)
+# 3. Build shared packages (can be skipped - npm run dev does it automatically)
 npm run build:packages
 
-# 4. Set up authentication (skip if using Claude Code CLI)
-# If using Claude Code CLI: credentials are detected automatically
-# If using API key directly, choose one method:
-
-# Option A: Environment variable
-export ANTHROPIC_API_KEY="sk-ant-..."
-
-# Option B: Create .env file in project root
-echo "ANTHROPIC_API_KEY=sk-ant-..." > .env
-
-# 5. Start Automaker (interactive launcher)
+# 4. Start Automaker
 npm run dev
 # Choose between:
 #   1. Web Application (browser at localhost:3007)
 #   2. Desktop Application (Electron - recommended)
 ```
 
-**Note:** The `npm run dev` command will:
+**Authentication Setup:** On first run, Automaker will automatically show a setup wizard where you can configure authentication. You can choose to:
 
-- Check for dependencies and install if needed
-- Install Playwright browsers for E2E tests
-- Kill any processes on ports 3007/3008
-- Present an interactive menu to choose your run mode
+- Use **Claude Code CLI** (recommended) - Automaker will detect your CLI credentials automatically
+- Enter an **API key** directly in the wizard
+
+If you prefer to set up authentication before running (e.g., for headless deployments or CI/CD), you can set it manually:
+
+```bash
+# Option A: Environment variable
+export ANTHROPIC_API_KEY="sk-ant-..."
+
+# Option B: Create .env file in project root
+echo "ANTHROPIC_API_KEY=sk-ant-..." > .env
+```
+
+**For Development:** `npm run dev` starts the development server with Vite live reload and hot module replacement for fast refresh and instant updates as you make changes.
 
 ## How to Run
 
@@ -186,9 +186,6 @@ npm run dev:web
 ```bash
 # Build for web deployment (uses Vite)
 npm run build
-
-# Run production build
-npm run start
 ```
 
 #### Desktop Application

TODO.md

@@ -0,0 +1,17 @@
+# Bugs
+
+- Setting the default model does not seem like it works.
+
+# UX
+
+- Consolidate all models to a single place in the settings instead of having AI profiles and all this other stuff
+- Simplify the create feature modal. It should just be one page. I don't need nessa tabs and all these nested buttons. It's too complex.
+- added to do's list checkbox directly into the card so as it's going through if there's any to do items we can see those update live
+- When the feature is done, I want to see a summary of the LLM. That's the first thing I should see when I double click the card.
+- I went away to mass edit all my features. For example, when I created a new project, it added auto testing on every single feature card. Now I have to manually go through one by one and change those. Have a way to mass edit those, the configuration of all them.
+- Double check and debug if there's memory leaks. It seems like the memory of automaker grows like 3 gigabytes. It's 5gb right now and I'm running three different cursor cli features implementing at the same time.
+- Typing in the text area of the plan mode was super laggy.
+- When I have a bunch of features running at the same time, it seems like I cannot edit the features in the backlog. Like they don't persist their file changes and I think this is because of the secure FS file has an internal queue to prevent hitting that file open write limit. We may have to reconsider refactoring away from file system and do Postgres or SQLite or something.
+- modals are not scrollable if height of the screen is small enough
+- and the Agent Runner add an archival button for the new sessions.
+- investigate a potential issue with the feature cards not refreshing. I see a lock icon on the feature card But it doesn't go away until I open the card and edit it and I turn the testing mode off. I think there's like a refresh sync issue.

apps/server/.env.example

@@ -8,6 +8,20 @@
 # Your Anthropic API key for Claude models
 ANTHROPIC_API_KEY=sk-ant-...
 
+# ============================================
+# OPTIONAL - Additional API Keys
+# ============================================
+
+# OpenAI API key for Codex/GPT models
+OPENAI_API_KEY=sk-...
+
+# Cursor API key for Cursor models
+CURSOR_API_KEY=...
+
+# OAuth credentials for CLI authentication (extracted automatically)
+CLAUDE_OAUTH_CREDENTIALS=
+CURSOR_AUTH_TOKEN=
+
 # ============================================
 # OPTIONAL - Security
 # ============================================
@@ -48,3 +62,15 @@ TERMINAL_ENABLED=true
 TERMINAL_PASSWORD=
 
 ENABLE_REQUEST_LOGGING=false
+
+# ============================================
+# OPTIONAL - Debugging
+# ============================================
+
+# Enable raw output logging for agent streams (default: false)
+# When enabled, saves unprocessed stream events to raw-output.jsonl
+# in each feature's directory (.automaker/features/{id}/raw-output.jsonl)
+# Useful for debugging provider streaming issues, improving log parsing,
+# or analyzing how different providers (Claude, Cursor) stream responses
+# Note: This adds disk I/O overhead, only enable when debugging
+AUTOMAKER_DEBUG_RAW_OUTPUT=false

apps/server/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@automaker/server",
-  "version": "0.7.3",
+  "version": "0.10.0",
   "description": "Backend server for Automaker - provides API for both web and Electron modes",
   "author": "AutoMaker Team",
   "license": "SEE LICENSE IN LICENSE",
@@ -32,7 +32,8 @@
     "@automaker/prompts": "1.0.0",
     "@automaker/types": "1.0.0",
     "@automaker/utils": "1.0.0",
-    "@modelcontextprotocol/sdk": "1.25.1",
+    "@modelcontextprotocol/sdk": "1.25.2",
+    "@openai/codex-sdk": "^0.77.0",
     "cookie-parser": "1.4.7",
     "cors": "2.8.5",
     "dotenv": "17.2.3",

apps/server/src/index.ts

@@ -17,6 +17,9 @@ import dotenv from 'dotenv';
 
 import { createEventEmitter, type EventEmitter } from './lib/events.js';
 import { initAllowedPaths } from '@automaker/platform';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('Server');
 import { authMiddleware, validateWsConnectionToken, checkRawAuthentication } from './lib/auth.js';
 import { requireJsonContentType } from './middleware/require-json-content-type.js';
 import { createAuthRoutes } from './routes/auth/index.js';
@@ -50,6 +53,10 @@ import { SettingsService } from './services/settings-service.js';
 import { createSpecRegenerationRoutes } from './routes/app-spec/index.js';
 import { createClaudeRoutes } from './routes/claude/index.js';
 import { ClaudeUsageService } from './services/claude-usage-service.js';
+import { createCodexRoutes } from './routes/codex/index.js';
+import { CodexUsageService } from './services/codex-usage-service.js';
+import { CodexAppServerService } from './services/codex-app-server-service.js';
+import { CodexModelCacheService } from './services/codex-model-cache-service.js';
 import { createGitHubRoutes } from './routes/github/index.js';
 import { createContextRoutes } from './routes/context/index.js';
 import { createBacklogPlanRoutes } from './routes/backlog-plan/index.js';
@@ -58,6 +65,8 @@ import { createMCPRoutes } from './routes/mcp/index.js';
 import { MCPTestService } from './services/mcp-test-service.js';
 import { createPipelineRoutes } from './routes/pipeline/index.js';
 import { pipelineService } from './services/pipeline-service.js';
+import { createIdeationRoutes } from './routes/ideation/index.js';
+import { IdeationService } from './services/ideation-service.js';
 
 // Load environment variables
 dotenv.config();
@@ -70,7 +79,7 @@ const ENABLE_REQUEST_LOGGING = process.env.ENABLE_REQUEST_LOGGING !== 'false'; /
 const hasAnthropicKey = !!process.env.ANTHROPIC_API_KEY;
 
 if (!hasAnthropicKey) {
-  console.warn(`
+  logger.warn(`
 ╔═══════════════════════════════════════════════════════════════════════╗
 ║  ⚠️  WARNING: No Claude authentication configured                      ║
 ║                                                                       ║
@@ -83,7 +92,7 @@ if (!hasAnthropicKey) {
 ╚═══════════════════════════════════════════════════════════════════════╝
 `);
 } else {
-  console.log('[Server] ✓ ANTHROPIC_API_KEY detected (API key auth)');
+  logger.info('✓ ANTHROPIC_API_KEY detected (API key auth)');
 }
 
 // Initialize security
@@ -161,12 +170,21 @@ const agentService = new AgentService(DATA_DIR, events, settingsService);
 const featureLoader = new FeatureLoader();
 const autoModeService = new AutoModeService(events, settingsService);
 const claudeUsageService = new ClaudeUsageService();
+const codexAppServerService = new CodexAppServerService();
+const codexModelCacheService = new CodexModelCacheService(DATA_DIR, codexAppServerService);
+const codexUsageService = new CodexUsageService(codexAppServerService);
 const mcpTestService = new MCPTestService(settingsService);
+const ideationService = new IdeationService(events, settingsService, featureLoader);
 
 // Initialize services
 (async () => {
   await agentService.initialize();
-  console.log('[Server] Agent service initialized');
+  logger.info('Agent service initialized');
+
+  // Bootstrap Codex model cache in background (don't block server startup)
+  void codexModelCacheService.getModels().catch((err) => {
+    logger.error('Failed to bootstrap Codex model cache:', err);
+  });
 })();
 
 // Run stale validation cleanup every hour to prevent memory leaks from crashed validations
@@ -174,7 +192,7 @@ const VALIDATION_CLEANUP_INTERVAL_MS = 60 * 60 * 1000; // 1 hour
 setInterval(() => {
   const cleaned = cleanupStaleValidations();
   if (cleaned > 0) {
-    console.log(`[Server] Cleaned up ${cleaned} stale validation entries`);
+    logger.info(`Cleaned up ${cleaned} stale validation entries`);
   }
 }, VALIDATION_CLEANUP_INTERVAL_MS);
 
@@ -182,9 +200,10 @@ setInterval(() => {
 // This helps prevent CSRF and content-type confusion attacks
 app.use('/api', requireJsonContentType);
 
-// Mount API routes - health and auth are unauthenticated
+// Mount API routes - health, auth, and setup are unauthenticated
 app.use('/api/health', createHealthRoutes());
 app.use('/api/auth', createAuthRoutes());
+app.use('/api/setup', createSetupRoutes());
 
 // Apply authentication to all other routes
 app.use('/api', authMiddleware);
@@ -198,9 +217,8 @@ app.use('/api/sessions', createSessionsRoutes(agentService));
 app.use('/api/features', createFeaturesRoutes(featureLoader));
 app.use('/api/auto-mode', createAutoModeRoutes(autoModeService));
 app.use('/api/enhance-prompt', createEnhancePromptRoutes(settingsService));
-app.use('/api/worktree', createWorktreeRoutes());
+app.use('/api/worktree', createWorktreeRoutes(events));
 app.use('/api/git', createGitRoutes());
-app.use('/api/setup', createSetupRoutes());
 app.use('/api/suggestions', createSuggestionsRoutes(events, settingsService));
 app.use('/api/models', createModelsRoutes());
 app.use('/api/spec-regeneration', createSpecRegenerationRoutes(events, settingsService));
@@ -210,11 +228,13 @@ app.use('/api/templates', createTemplatesRoutes());
 app.use('/api/terminal', createTerminalRoutes());
 app.use('/api/settings', createSettingsRoutes(settingsService));
 app.use('/api/claude', createClaudeRoutes(claudeUsageService));
+app.use('/api/codex', createCodexRoutes(codexUsageService, codexModelCacheService));
 app.use('/api/github', createGitHubRoutes(events, settingsService));
 app.use('/api/context', createContextRoutes(settingsService));
 app.use('/api/backlog-plan', createBacklogPlanRoutes(events, settingsService));
 app.use('/api/mcp', createMCPRoutes(mcpTestService));
 app.use('/api/pipeline', createPipelineRoutes(pipelineService));
+app.use('/api/ideation', createIdeationRoutes(events, ideationService, featureLoader));
 
 // Create HTTP server
 const server = createServer(app);
@@ -267,7 +287,7 @@ server.on('upgrade', (request, socket, head) => {
 
   // Authenticate all WebSocket connections
   if (!authenticateWebSocket(request)) {
-    console.log('[WebSocket] Authentication failed, rejecting connection');
+    logger.info('Authentication failed, rejecting connection');
     socket.write('HTTP/1.1 401 Unauthorized\r\n\r\n');
     socket.destroy();
     return;
@@ -288,11 +308,11 @@ server.on('upgrade', (request, socket, head) => {
 
 // Events WebSocket connection handler
 wss.on('connection', (ws: WebSocket) => {
-  console.log('[WebSocket] Client connected, ready state:', ws.readyState);
+  logger.info('Client connected, ready state:', ws.readyState);
 
   // Subscribe to all events and forward to this client
   const unsubscribe = events.subscribe((type, payload) => {
-    console.log('[WebSocket] Event received:', {
+    logger.info('Event received:', {
       type,
       hasPayload: !!payload,
       payloadKeys: payload ? Object.keys(payload) : [],
@@ -302,27 +322,24 @@ wss.on('connection', (ws: WebSocket) => {
 
     if (ws.readyState === WebSocket.OPEN) {
       const message = JSON.stringify({ type, payload });
-      console.log('[WebSocket] Sending event to client:', {
+      logger.info('Sending event to client:', {
         type,
         messageLength: message.length,
         sessionId: (payload as any)?.sessionId,
       });
       ws.send(message);
     } else {
-      console.log(
-        '[WebSocket] WARNING: Cannot send event, WebSocket not open. ReadyState:',
-        ws.readyState
-      );
+      logger.info('WARNING: Cannot send event, WebSocket not open. ReadyState:', ws.readyState);
     }
   });
 
   ws.on('close', () => {
-    console.log('[WebSocket] Client disconnected');
+    logger.info('Client disconnected');
     unsubscribe();
   });
 
   ws.on('error', (error) => {
-    console.error('[WebSocket] ERROR:', error);
+    logger.error('ERROR:', error);
     unsubscribe();
   });
 });
@@ -349,24 +366,24 @@ terminalWss.on('connection', (ws: WebSocket, req: import('http').IncomingMessage
   const sessionId = url.searchParams.get('sessionId');
   const token = url.searchParams.get('token');
 
-  console.log(`[Terminal WS] Connection attempt for session: ${sessionId}`);
+  logger.info(`Connection attempt for session: ${sessionId}`);
 
   // Check if terminal is enabled
   if (!isTerminalEnabled()) {
-    console.log('[Terminal WS] Terminal is disabled');
+    logger.info('Terminal is disabled');
     ws.close(4003, 'Terminal access is disabled');
     return;
   }
 
   // Validate token if password is required
   if (isTerminalPasswordRequired() && !validateTerminalToken(token || undefined)) {
-    console.log('[Terminal WS] Invalid or missing token');
+    logger.info('Invalid or missing token');
     ws.close(4001, 'Authentication required');
     return;
   }
 
   if (!sessionId) {
-    console.log('[Terminal WS] No session ID provided');
+    logger.info('No session ID provided');
     ws.close(4002, 'Session ID required');
     return;
   }
@@ -374,12 +391,12 @@ terminalWss.on('connection', (ws: WebSocket, req: import('http').IncomingMessage
   // Check if session exists
   const session = terminalService.getSession(sessionId);
   if (!session) {
-    console.log(`[Terminal WS] Session ${sessionId} not found`);
+    logger.info(`Session ${sessionId} not found`);
     ws.close(4004, 'Session not found');
     return;
   }
 
-  console.log(`[Terminal WS] Client connected to session ${sessionId}`);
+  logger.info(`Client connected to session ${sessionId}`);
 
   // Track this connection
   if (!terminalConnections.has(sessionId)) {
@@ -495,15 +512,15 @@ terminalWss.on('connection', (ws: WebSocket, req: import('http').IncomingMessage
           break;
 
         default:
-          console.warn(`[Terminal WS] Unknown message type: ${msg.type}`);
+          logger.warn(`Unknown message type: ${msg.type}`);
       }
     } catch (error) {
-      console.error('[Terminal WS] Error processing message:', error);
+      logger.error('Error processing message:', error);
     }
   });
 
   ws.on('close', () => {
-    console.log(`[Terminal WS] Client disconnected from session ${sessionId}`);
+    logger.info(`Client disconnected from session ${sessionId}`);
     unsubscribeData();
     unsubscribeExit();
 
@@ -522,7 +539,7 @@ terminalWss.on('connection', (ws: WebSocket, req: import('http').IncomingMessage
   });
 
   ws.on('error', (error) => {
-    console.error(`[Terminal WS] Error on session ${sessionId}:`, error);
+    logger.error(`Error on session ${sessionId}:`, error);
     unsubscribeData();
     unsubscribeExit();
   });
@@ -537,7 +554,7 @@ const startServer = (port: number) => {
         : 'enabled'
       : 'disabled';
     const portStr = port.toString().padEnd(4);
-    console.log(`
+    logger.info(`
 ╔═══════════════════════════════════════════════════════╗
 ║           Automaker Backend Server                    ║
 ╠═══════════════════════════════════════════════════════╣
@@ -552,7 +569,7 @@ const startServer = (port: number) => {
 
   server.on('error', (error: NodeJS.ErrnoException) => {
     if (error.code === 'EADDRINUSE') {
-      console.error(`
+      logger.error(`
 ╔═══════════════════════════════════════════════════════╗
 ║  ❌ ERROR: Port ${port} is already in use              ║
 ╠═══════════════════════════════════════════════════════╣
@@ -572,7 +589,7 @@ const startServer = (port: number) => {
 `);
       process.exit(1);
     } else {
-      console.error('[Server] Error starting server:', error);
+      logger.error('Error starting server:', error);
       process.exit(1);
     }
   });
@@ -580,21 +597,41 @@ const startServer = (port: number) => {
 
 startServer(PORT);
 
+// Global error handlers to prevent crashes from uncaught errors
+process.on('unhandledRejection', (reason: unknown, _promise: Promise<unknown>) => {
+  logger.error('Unhandled Promise Rejection:', {
+    reason: reason instanceof Error ? reason.message : String(reason),
+    stack: reason instanceof Error ? reason.stack : undefined,
+  });
+  // Don't exit - log the error and continue running
+  // This prevents the server from crashing due to unhandled rejections
+});
+
+process.on('uncaughtException', (error: Error) => {
+  logger.error('Uncaught Exception:', {
+    message: error.message,
+    stack: error.stack,
+  });
+  // Exit on uncaught exceptions to prevent undefined behavior
+  // The process is in an unknown state after an uncaught exception
+  process.exit(1);
+});
+
 // Graceful shutdown
 process.on('SIGTERM', () => {
-  console.log('SIGTERM received, shutting down...');
+  logger.info('SIGTERM received, shutting down...');
   terminalService.cleanup();
   server.close(() => {
-    console.log('Server closed');
+    logger.info('Server closed');
     process.exit(0);
   });
 });
 
 process.on('SIGINT', () => {
-  console.log('SIGINT received, shutting down...');
+  logger.info('SIGINT received, shutting down...');
   terminalService.cleanup();
   server.close(() => {
-    console.log('Server closed');
+    logger.info('Server closed');
     process.exit(0);
   });
 });

apps/server/src/lib/agent-discovery.ts

@@ -0,0 +1,257 @@
+/**
+ * Agent Discovery - Scans filesystem for AGENT.md files
+ *
+ * Discovers agents from:
+ * - ~/.claude/agents/ (user-level, global)
+ * - .claude/agents/ (project-level)
+ *
+ * Similar to Skills, but for custom subagents defined in AGENT.md files.
+ */
+
+import path from 'path';
+import os from 'os';
+import { createLogger } from '@automaker/utils';
+import { secureFs, systemPaths } from '@automaker/platform';
+import type { AgentDefinition } from '@automaker/types';
+
+const logger = createLogger('AgentDiscovery');
+
+export interface FilesystemAgent {
+  name: string; // Directory name (e.g., 'code-reviewer')
+  definition: AgentDefinition;
+  source: 'user' | 'project';
+  filePath: string; // Full path to AGENT.md
+}
+
+/**
+ * Parse agent content string into AgentDefinition
+ * Format:
+ * ---
+ * name: agent-name  # Optional
+ * description: When to use this agent
+ * tools: tool1, tool2, tool3  # Optional (comma or space separated list)
+ * model: sonnet  # Optional: sonnet, opus, haiku
+ * ---
+ * System prompt content here...
+ */
+function parseAgentContent(content: string, filePath: string): AgentDefinition | null {
+  // Extract frontmatter
+  const frontmatterMatch = content.match(/^---\n([\s\S]*?)\n---\n([\s\S]*)$/);
+  if (!frontmatterMatch) {
+    logger.warn(`Invalid agent file format (missing frontmatter): ${filePath}`);
+    return null;
+  }
+
+  const [, frontmatter, prompt] = frontmatterMatch;
+
+  // Parse description (required)
+  const description = frontmatter.match(/description:\s*(.+)/)?.[1]?.trim();
+  if (!description) {
+    logger.warn(`Missing description in agent file: ${filePath}`);
+    return null;
+  }
+
+  // Parse tools (optional) - supports both comma-separated and space-separated
+  const toolsMatch = frontmatter.match(/tools:\s*(.+)/);
+  const tools = toolsMatch
+    ? toolsMatch[1]
+        .split(/[,\s]+/) // Split by comma or whitespace
+        .map((t) => t.trim())
+        .filter((t) => t && t !== '')
+    : undefined;
+
+  // Parse model (optional) - validate against allowed values
+  const modelMatch = frontmatter.match(/model:\s*(\w+)/);
+  const modelValue = modelMatch?.[1]?.trim();
+  const validModels = ['sonnet', 'opus', 'haiku', 'inherit'] as const;
+  const model =
+    modelValue && validModels.includes(modelValue as (typeof validModels)[number])
+      ? (modelValue as 'sonnet' | 'opus' | 'haiku' | 'inherit')
+      : undefined;
+
+  if (modelValue && !model) {
+    logger.warn(
+      `Invalid model "${modelValue}" in agent file: ${filePath}. Expected one of: ${validModels.join(', ')}`
+    );
+  }
+
+  return {
+    description,
+    prompt: prompt.trim(),
+    tools,
+    model,
+  };
+}
+
+/**
+ * Directory entry with type information
+ */
+interface DirEntry {
+  name: string;
+  isFile: boolean;
+  isDirectory: boolean;
+}
+
+/**
+ * Filesystem adapter interface for abstracting systemPaths vs secureFs
+ */
+interface FsAdapter {
+  exists: (filePath: string) => Promise<boolean>;
+  readdir: (dirPath: string) => Promise<DirEntry[]>;
+  readFile: (filePath: string) => Promise<string>;
+}
+
+/**
+ * Create a filesystem adapter for system paths (user directory)
+ */
+function createSystemPathAdapter(): FsAdapter {
+  return {
+    exists: (filePath) => Promise.resolve(systemPaths.systemPathExists(filePath)),
+    readdir: async (dirPath) => {
+      const entryNames = await systemPaths.systemPathReaddir(dirPath);
+      const entries: DirEntry[] = [];
+      for (const name of entryNames) {
+        const stat = await systemPaths.systemPathStat(path.join(dirPath, name));
+        entries.push({
+          name,
+          isFile: stat.isFile(),
+          isDirectory: stat.isDirectory(),
+        });
+      }
+      return entries;
+    },
+    readFile: (filePath) => systemPaths.systemPathReadFile(filePath, 'utf-8') as Promise<string>,
+  };
+}
+
+/**
+ * Create a filesystem adapter for project paths (secureFs)
+ */
+function createSecureFsAdapter(): FsAdapter {
+  return {
+    exists: (filePath) =>
+      secureFs
+        .access(filePath)
+        .then(() => true)
+        .catch(() => false),
+    readdir: async (dirPath) => {
+      const entries = await secureFs.readdir(dirPath, { withFileTypes: true });
+      return entries.map((entry) => ({
+        name: entry.name,
+        isFile: entry.isFile(),
+        isDirectory: entry.isDirectory(),
+      }));
+    },
+    readFile: (filePath) => secureFs.readFile(filePath, 'utf-8') as Promise<string>,
+  };
+}
+
+/**
+ * Parse agent file using the provided filesystem adapter
+ */
+async function parseAgentFileWithAdapter(
+  filePath: string,
+  fsAdapter: FsAdapter
+): Promise<AgentDefinition | null> {
+  try {
+    const content = await fsAdapter.readFile(filePath);
+    return parseAgentContent(content, filePath);
+  } catch (error) {
+    logger.error(`Failed to parse agent file: ${filePath}`, error);
+    return null;
+  }
+}
+
+/**
+ * Scan a directory for agent .md files
+ * Agents can be in two formats:
+ * 1. Flat: agent-name.md (file directly in agents/)
+ * 2. Subdirectory: agent-name/AGENT.md (folder + file, similar to Skills)
+ */
+async function scanAgentsDirectory(
+  baseDir: string,
+  source: 'user' | 'project'
+): Promise<FilesystemAgent[]> {
+  const agents: FilesystemAgent[] = [];
+  const fsAdapter = source === 'user' ? createSystemPathAdapter() : createSecureFsAdapter();
+
+  try {
+    // Check if directory exists
+    const exists = await fsAdapter.exists(baseDir);
+    if (!exists) {
+      logger.debug(`Directory does not exist: ${baseDir}`);
+      return agents;
+    }
+
+    // Read all entries in the directory
+    const entries = await fsAdapter.readdir(baseDir);
+
+    for (const entry of entries) {
+      // Check for flat .md file format (agent-name.md)
+      if (entry.isFile && entry.name.endsWith('.md')) {
+        const agentName = entry.name.slice(0, -3); // Remove .md extension
+        const agentFilePath = path.join(baseDir, entry.name);
+        const definition = await parseAgentFileWithAdapter(agentFilePath, fsAdapter);
+        if (definition) {
+          agents.push({
+            name: agentName,
+            definition,
+            source,
+            filePath: agentFilePath,
+          });
+          logger.debug(`Discovered ${source} agent (flat): ${agentName}`);
+        }
+      }
+      // Check for subdirectory format (agent-name/AGENT.md)
+      else if (entry.isDirectory) {
+        const agentFilePath = path.join(baseDir, entry.name, 'AGENT.md');
+        const agentFileExists = await fsAdapter.exists(agentFilePath);
+
+        if (agentFileExists) {
+          const definition = await parseAgentFileWithAdapter(agentFilePath, fsAdapter);
+          if (definition) {
+            agents.push({
+              name: entry.name,
+              definition,
+              source,
+              filePath: agentFilePath,
+            });
+            logger.debug(`Discovered ${source} agent (subdirectory): ${entry.name}`);
+          }
+        }
+      }
+    }
+  } catch (error) {
+    logger.error(`Failed to scan agents directory: ${baseDir}`, error);
+  }
+
+  return agents;
+}
+
+/**
+ * Discover all filesystem-based agents from user and project sources
+ */
+export async function discoverFilesystemAgents(
+  projectPath?: string,
+  sources: Array<'user' | 'project'> = ['user', 'project']
+): Promise<FilesystemAgent[]> {
+  const agents: FilesystemAgent[] = [];
+
+  // Discover user-level agents from ~/.claude/agents/
+  if (sources.includes('user')) {
+    const userAgentsDir = path.join(os.homedir(), '.claude', 'agents');
+    const userAgents = await scanAgentsDirectory(userAgentsDir, 'user');
+    agents.push(...userAgents);
+    logger.info(`Discovered ${userAgents.length} user-level agents from ${userAgentsDir}`);
+  }
+
+  // Discover project-level agents from .claude/agents/
+  if (sources.includes('project') && projectPath) {
+    const projectAgentsDir = path.join(projectPath, '.claude', 'agents');
+    const projectAgents = await scanAgentsDirectory(projectAgentsDir, 'project');
+    agents.push(...projectAgents);
+    logger.info(`Discovered ${projectAgents.length} project-level agents from ${projectAgentsDir}`);
+  }
+
+  return agents;
+}

apps/server/src/lib/auth-utils.ts

@@ -0,0 +1,263 @@
+/**
+ * Secure authentication utilities that avoid environment variable race conditions
+ */
+
+import { spawn } from 'child_process';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('AuthUtils');
+
+export interface SecureAuthEnv {
+  [key: string]: string | undefined;
+}
+
+export interface AuthValidationResult {
+  isValid: boolean;
+  error?: string;
+  normalizedKey?: string;
+}
+
+/**
+ * Validates API key format without modifying process.env
+ */
+export function validateApiKey(
+  key: string,
+  provider: 'anthropic' | 'openai' | 'cursor'
+): AuthValidationResult {
+  if (!key || typeof key !== 'string' || key.trim().length === 0) {
+    return { isValid: false, error: 'API key is required' };
+  }
+
+  const trimmedKey = key.trim();
+
+  switch (provider) {
+    case 'anthropic':
+      if (!trimmedKey.startsWith('sk-ant-')) {
+        return {
+          isValid: false,
+          error: 'Invalid Anthropic API key format. Should start with "sk-ant-"',
+        };
+      }
+      if (trimmedKey.length < 20) {
+        return { isValid: false, error: 'Anthropic API key too short' };
+      }
+      break;
+
+    case 'openai':
+      if (!trimmedKey.startsWith('sk-')) {
+        return { isValid: false, error: 'Invalid OpenAI API key format. Should start with "sk-"' };
+      }
+      if (trimmedKey.length < 20) {
+        return { isValid: false, error: 'OpenAI API key too short' };
+      }
+      break;
+
+    case 'cursor':
+      // Cursor API keys might have different format
+      if (trimmedKey.length < 10) {
+        return { isValid: false, error: 'Cursor API key too short' };
+      }
+      break;
+  }
+
+  return { isValid: true, normalizedKey: trimmedKey };
+}
+
+/**
+ * Creates a secure environment object for authentication testing
+ * without modifying the global process.env
+ */
+export function createSecureAuthEnv(
+  authMethod: 'cli' | 'api_key',
+  apiKey?: string,
+  provider: 'anthropic' | 'openai' | 'cursor' = 'anthropic'
+): SecureAuthEnv {
+  const env: SecureAuthEnv = { ...process.env };
+
+  if (authMethod === 'cli') {
+    // For CLI auth, remove the API key to force CLI authentication
+    const envKey = provider === 'openai' ? 'OPENAI_API_KEY' : 'ANTHROPIC_API_KEY';
+    delete env[envKey];
+  } else if (authMethod === 'api_key' && apiKey) {
+    // For API key auth, validate and set the provided key
+    const validation = validateApiKey(apiKey, provider);
+    if (!validation.isValid) {
+      throw new Error(validation.error);
+    }
+    const envKey = provider === 'openai' ? 'OPENAI_API_KEY' : 'ANTHROPIC_API_KEY';
+    env[envKey] = validation.normalizedKey;
+  }
+
+  return env;
+}
+
+/**
+ * Creates a temporary environment override for the current process
+ * WARNING: This should only be used in isolated contexts and immediately cleaned up
+ */
+export function createTempEnvOverride(authEnv: SecureAuthEnv): () => void {
+  const originalEnv = { ...process.env };
+
+  // Apply the auth environment
+  Object.assign(process.env, authEnv);
+
+  // Return cleanup function
+  return () => {
+    // Restore original environment
+    Object.keys(process.env).forEach((key) => {
+      if (!(key in originalEnv)) {
+        delete process.env[key];
+      }
+    });
+    Object.assign(process.env, originalEnv);
+  };
+}
+
+/**
+ * Spawns a process with secure environment isolation
+ */
+export function spawnSecureAuth(
+  command: string,
+  args: string[],
+  authEnv: SecureAuthEnv,
+  options: {
+    cwd?: string;
+    timeout?: number;
+  } = {}
+): Promise<{ stdout: string; stderr: string; exitCode: number | null }> {
+  return new Promise((resolve, reject) => {
+    const { cwd = process.cwd(), timeout = 30000 } = options;
+
+    logger.debug(`Spawning secure auth process: ${command} ${args.join(' ')}`);
+
+    const child = spawn(command, args, {
+      cwd,
+      env: authEnv,
+      stdio: 'pipe',
+      shell: false,
+    });
+
+    let stdout = '';
+    let stderr = '';
+    let isResolved = false;
+
+    const timeoutId = setTimeout(() => {
+      if (!isResolved) {
+        child.kill('SIGTERM');
+        isResolved = true;
+        reject(new Error('Authentication process timed out'));
+      }
+    }, timeout);
+
+    child.stdout?.on('data', (data) => {
+      stdout += data.toString();
+    });
+
+    child.stderr?.on('data', (data) => {
+      stderr += data.toString();
+    });
+
+    child.on('close', (code) => {
+      clearTimeout(timeoutId);
+      if (!isResolved) {
+        isResolved = true;
+        resolve({ stdout, stderr, exitCode: code });
+      }
+    });
+
+    child.on('error', (error) => {
+      clearTimeout(timeoutId);
+      if (!isResolved) {
+        isResolved = true;
+        reject(error);
+      }
+    });
+  });
+}
+
+/**
+ * Safely extracts environment variable without race conditions
+ */
+export function safeGetEnv(key: string): string | undefined {
+  return process.env[key];
+}
+
+/**
+ * Checks if an environment variable would be modified without actually modifying it
+ */
+export function wouldModifyEnv(key: string, newValue: string): boolean {
+  const currentValue = safeGetEnv(key);
+  return currentValue !== newValue;
+}
+
+/**
+ * Secure auth session management
+ */
+export class AuthSessionManager {
+  private static activeSessions = new Map<string, SecureAuthEnv>();
+
+  static createSession(
+    sessionId: string,
+    authMethod: 'cli' | 'api_key',
+    apiKey?: string,
+    provider: 'anthropic' | 'openai' | 'cursor' = 'anthropic'
+  ): SecureAuthEnv {
+    const env = createSecureAuthEnv(authMethod, apiKey, provider);
+    this.activeSessions.set(sessionId, env);
+    return env;
+  }
+
+  static getSession(sessionId: string): SecureAuthEnv | undefined {
+    return this.activeSessions.get(sessionId);
+  }
+
+  static destroySession(sessionId: string): void {
+    this.activeSessions.delete(sessionId);
+  }
+
+  static cleanup(): void {
+    this.activeSessions.clear();
+  }
+}
+
+/**
+ * Rate limiting for auth attempts to prevent abuse
+ */
+export class AuthRateLimiter {
+  private attempts = new Map<string, { count: number; lastAttempt: number }>();
+
+  constructor(
+    private maxAttempts = 5,
+    private windowMs = 60000
+  ) {}
+
+  canAttempt(identifier: string): boolean {
+    const now = Date.now();
+    const record = this.attempts.get(identifier);
+
+    if (!record || now - record.lastAttempt > this.windowMs) {
+      this.attempts.set(identifier, { count: 1, lastAttempt: now });
+      return true;
+    }
+
+    if (record.count >= this.maxAttempts) {
+      return false;
+    }
+
+    record.count++;
+    record.lastAttempt = now;
+    return true;
+  }
+
+  getRemainingAttempts(identifier: string): number {
+    const record = this.attempts.get(identifier);
+    if (!record) return this.maxAttempts;
+    return Math.max(0, this.maxAttempts - record.count);
+  }
+
+  getResetTime(identifier: string): Date | null {
+    const record = this.attempts.get(identifier);
+    if (!record) return null;
+    return new Date(record.lastAttempt + this.windowMs);
+  }
+}

apps/server/src/lib/auth.ts

@@ -12,6 +12,9 @@ import type { Request, Response, NextFunction } from 'express';
 import crypto from 'crypto';
 import path from 'path';
 import * as secureFs from './secure-fs.js';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('Auth');
 
 const DATA_DIR = process.env.DATA_DIR || './data';
 const API_KEY_FILE = path.join(DATA_DIR, '.api-key');
@@ -61,11 +64,11 @@ function loadSessions(): void {
       }
 
       if (loadedCount > 0 || expiredCount > 0) {
-        console.log(`[Auth] Loaded ${loadedCount} sessions (${expiredCount} expired)`);
+        logger.info(`Loaded ${loadedCount} sessions (${expiredCount} expired)`);
       }
     }
   } catch (error) {
-    console.warn('[Auth] Error loading sessions:', error);
+    logger.warn('Error loading sessions:', error);
   }
 }
 
@@ -81,7 +84,7 @@ async function saveSessions(): Promise<void> {
       mode: 0o600,
     });
   } catch (error) {
-    console.error('[Auth] Failed to save sessions:', error);
+    logger.error('Failed to save sessions:', error);
   }
 }
 
@@ -95,7 +98,7 @@ loadSessions();
 function ensureApiKey(): string {
   // First check environment variable (Electron passes it this way)
   if (process.env.AUTOMAKER_API_KEY) {
-    console.log('[Auth] Using API key from environment variable');
+    logger.info('Using API key from environment variable');
     return process.env.AUTOMAKER_API_KEY;
   }
 
@@ -104,12 +107,12 @@ function ensureApiKey(): string {
     if (secureFs.existsSync(API_KEY_FILE)) {
       const key = (secureFs.readFileSync(API_KEY_FILE, 'utf-8') as string).trim();
       if (key) {
-        console.log('[Auth] Loaded API key from file');
+        logger.info('Loaded API key from file');
         return key;
       }
     }
   } catch (error) {
-    console.warn('[Auth] Error reading API key file:', error);
+    logger.warn('Error reading API key file:', error);
   }
 
   // Generate new key
@@ -117,9 +120,9 @@ function ensureApiKey(): string {
   try {
     secureFs.mkdirSync(path.dirname(API_KEY_FILE), { recursive: true });
     secureFs.writeFileSync(API_KEY_FILE, newKey, { encoding: 'utf-8', mode: 0o600 });
-    console.log('[Auth] Generated new API key');
+    logger.info('Generated new API key');
   } catch (error) {
-    console.error('[Auth] Failed to save API key:', error);
+    logger.error('Failed to save API key:', error);
   }
   return newKey;
 }
@@ -129,7 +132,7 @@ const API_KEY = ensureApiKey();
 
 // Print API key to console for web mode users (unless suppressed for production logging)
 if (process.env.AUTOMAKER_HIDE_API_KEY !== 'true') {
-  console.log(`
+  logger.info(`
 ╔═══════════════════════════════════════════════════════════════════════╗
 ║  🔐 API Key for Web Mode Authentication                               ║
 ╠═══════════════════════════════════════════════════════════════════════╣
@@ -142,7 +145,7 @@ if (process.env.AUTOMAKER_HIDE_API_KEY !== 'true') {
 ╚═══════════════════════════════════════════════════════════════════════╝
 `);
 } else {
-  console.log('[Auth] API key banner hidden (AUTOMAKER_HIDE_API_KEY=true)');
+  logger.info('API key banner hidden (AUTOMAKER_HIDE_API_KEY=true)');
 }
 
 /**
@@ -177,7 +180,7 @@ export function validateSession(token: string): boolean {
   if (Date.now() > session.expiresAt) {
     validSessions.delete(token);
     // Fire-and-forget: persist removal asynchronously
-    saveSessions().catch((err) => console.error('[Auth] Error saving sessions:', err));
+    saveSessions().catch((err) => logger.error('Error saving sessions:', err));
     return false;
   }
 
@@ -259,7 +262,7 @@ export function getSessionCookieOptions(): {
   return {
     httpOnly: true, // JavaScript cannot access this cookie
     secure: process.env.NODE_ENV === 'production', // HTTPS only in production
-    sameSite: 'strict', // Only sent for same-site requests (CSRF protection)
+    sameSite: 'lax', // Sent for same-site requests and top-level navigations, but not cross-origin fetch/XHR
     maxAge: SESSION_MAX_AGE_MS,
     path: '/',
   };

apps/server/src/lib/cli-detection.ts

@@ -0,0 +1,447 @@
+/**
+ * Unified CLI Detection Framework
+ *
+ * Provides consistent CLI detection and management across all providers
+ */
+
+import { spawn, execSync } from 'child_process';
+import * as fs from 'fs';
+import * as path from 'path';
+import * as os from 'os';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('CliDetection');
+
+export interface CliInfo {
+  name: string;
+  command: string;
+  version?: string;
+  path?: string;
+  installed: boolean;
+  authenticated: boolean;
+  authMethod: 'cli' | 'api_key' | 'none';
+  platform?: string;
+  architectures?: string[];
+}
+
+export interface CliDetectionOptions {
+  timeout?: number;
+  includeWsl?: boolean;
+  wslDistribution?: string;
+}
+
+export interface CliDetectionResult {
+  cli: CliInfo;
+  detected: boolean;
+  issues: string[];
+}
+
+export interface UnifiedCliDetection {
+  claude?: CliDetectionResult;
+  codex?: CliDetectionResult;
+  cursor?: CliDetectionResult;
+}
+
+/**
+ * CLI Configuration for different providers
+ */
+const CLI_CONFIGS = {
+  claude: {
+    name: 'Claude CLI',
+    commands: ['claude'],
+    versionArgs: ['--version'],
+    installCommands: {
+      darwin: 'brew install anthropics/claude/claude',
+      linux: 'curl -fsSL https://claude.ai/install.sh | sh',
+      win32: 'iwr https://claude.ai/install.ps1 -UseBasicParsing | iex',
+    },
+  },
+  codex: {
+    name: 'Codex CLI',
+    commands: ['codex', 'openai'],
+    versionArgs: ['--version'],
+    installCommands: {
+      darwin: 'npm install -g @openai/codex-cli',
+      linux: 'npm install -g @openai/codex-cli',
+      win32: 'npm install -g @openai/codex-cli',
+    },
+  },
+  cursor: {
+    name: 'Cursor CLI',
+    commands: ['cursor-agent', 'cursor'],
+    versionArgs: ['--version'],
+    installCommands: {
+      darwin: 'brew install cursor/cursor/cursor-agent',
+      linux: 'curl -fsSL https://cursor.sh/install.sh | sh',
+      win32: 'iwr https://cursor.sh/install.ps1 -UseBasicParsing | iex',
+    },
+  },
+} as const;
+
+/**
+ * Detect if a CLI is installed and available
+ */
+export async function detectCli(
+  provider: keyof typeof CLI_CONFIGS,
+  options: CliDetectionOptions = {}
+): Promise<CliDetectionResult> {
+  const config = CLI_CONFIGS[provider];
+  const { timeout = 5000, includeWsl = false, wslDistribution } = options;
+  const issues: string[] = [];
+
+  const cliInfo: CliInfo = {
+    name: config.name,
+    command: '',
+    installed: false,
+    authenticated: false,
+    authMethod: 'none',
+  };
+
+  try {
+    // Find the command in PATH
+    const command = await findCommand([...config.commands]);
+    if (command) {
+      cliInfo.command = command;
+    }
+
+    if (!cliInfo.command) {
+      issues.push(`${config.name} not found in PATH`);
+      return { cli: cliInfo, detected: false, issues };
+    }
+
+    cliInfo.path = cliInfo.command;
+    cliInfo.installed = true;
+
+    // Get version
+    try {
+      cliInfo.version = await getCliVersion(cliInfo.command, [...config.versionArgs], timeout);
+    } catch (error) {
+      issues.push(`Failed to get ${config.name} version: ${error}`);
+    }
+
+    // Check authentication
+    cliInfo.authMethod = await checkCliAuth(provider, cliInfo.command);
+    cliInfo.authenticated = cliInfo.authMethod !== 'none';
+
+    return { cli: cliInfo, detected: true, issues };
+  } catch (error) {
+    issues.push(`Error detecting ${config.name}: ${error}`);
+    return { cli: cliInfo, detected: false, issues };
+  }
+}
+
+/**
+ * Detect all CLIs in the system
+ */
+export async function detectAllCLis(
+  options: CliDetectionOptions = {}
+): Promise<UnifiedCliDetection> {
+  const results: UnifiedCliDetection = {};
+
+  // Detect all providers in parallel
+  const providers = Object.keys(CLI_CONFIGS) as Array<keyof typeof CLI_CONFIGS>;
+  const detectionPromises = providers.map(async (provider) => {
+    const result = await detectCli(provider, options);
+    return { provider, result };
+  });
+
+  const detections = await Promise.all(detectionPromises);
+
+  for (const { provider, result } of detections) {
+    results[provider] = result;
+  }
+
+  return results;
+}
+
+/**
+ * Find the first available command from a list of alternatives
+ */
+export async function findCommand(commands: string[]): Promise<string | null> {
+  for (const command of commands) {
+    try {
+      const whichCommand = process.platform === 'win32' ? 'where' : 'which';
+      const result = execSync(`${whichCommand} ${command}`, {
+        encoding: 'utf8',
+        timeout: 2000,
+      }).trim();
+
+      if (result) {
+        return result.split('\n')[0]; // Take first result on Windows
+      }
+    } catch {
+      // Command not found, try next
+    }
+  }
+  return null;
+}
+
+/**
+ * Get CLI version
+ */
+export async function getCliVersion(
+  command: string,
+  args: string[],
+  timeout: number = 5000
+): Promise<string> {
+  return new Promise((resolve, reject) => {
+    const child = spawn(command, args, {
+      stdio: 'pipe',
+      timeout,
+    });
+
+    let stdout = '';
+    let stderr = '';
+
+    child.stdout?.on('data', (data) => {
+      stdout += data.toString();
+    });
+
+    child.stderr?.on('data', (data) => {
+      stderr += data.toString();
+    });
+
+    child.on('close', (code) => {
+      if (code === 0 && stdout) {
+        resolve(stdout.trim());
+      } else if (stderr) {
+        reject(stderr.trim());
+      } else {
+        reject(`Command exited with code ${code}`);
+      }
+    });
+
+    child.on('error', reject);
+  });
+}
+
+/**
+ * Check authentication status for a CLI
+ */
+export async function checkCliAuth(
+  provider: keyof typeof CLI_CONFIGS,
+  command: string
+): Promise<'cli' | 'api_key' | 'none'> {
+  try {
+    switch (provider) {
+      case 'claude':
+        return await checkClaudeAuth(command);
+      case 'codex':
+        return await checkCodexAuth(command);
+      case 'cursor':
+        return await checkCursorAuth(command);
+      default:
+        return 'none';
+    }
+  } catch {
+    return 'none';
+  }
+}
+
+/**
+ * Check Claude CLI authentication
+ */
+async function checkClaudeAuth(command: string): Promise<'cli' | 'api_key' | 'none'> {
+  try {
+    // Check for environment variable
+    if (process.env.ANTHROPIC_API_KEY) {
+      return 'api_key';
+    }
+
+    // Try running a simple command to check CLI auth
+    const result = await getCliVersion(command, ['--version'], 3000);
+    if (result) {
+      return 'cli'; // If version works, assume CLI is authenticated
+    }
+  } catch {
+    // Version command might work even without auth, so we need a better check
+  }
+
+  // Try a more specific auth check
+  return new Promise((resolve) => {
+    const child = spawn(command, ['whoami'], {
+      stdio: 'pipe',
+      timeout: 3000,
+    });
+
+    let stdout = '';
+    let stderr = '';
+
+    child.stdout?.on('data', (data) => {
+      stdout += data.toString();
+    });
+
+    child.stderr?.on('data', (data) => {
+      stderr += data.toString();
+    });
+
+    child.on('close', (code) => {
+      if (code === 0 && stdout && !stderr.includes('not authenticated')) {
+        resolve('cli');
+      } else {
+        resolve('none');
+      }
+    });
+
+    child.on('error', () => {
+      resolve('none');
+    });
+  });
+}
+
+/**
+ * Check Codex CLI authentication
+ */
+async function checkCodexAuth(command: string): Promise<'cli' | 'api_key' | 'none'> {
+  // Check for environment variable
+  if (process.env.OPENAI_API_KEY) {
+    return 'api_key';
+  }
+
+  try {
+    // Try a simple auth check
+    const result = await getCliVersion(command, ['--version'], 3000);
+    if (result) {
+      return 'cli';
+    }
+  } catch {
+    // Version check failed
+  }
+
+  return 'none';
+}
+
+/**
+ * Check Cursor CLI authentication
+ */
+async function checkCursorAuth(command: string): Promise<'cli' | 'api_key' | 'none'> {
+  // Check for environment variable
+  if (process.env.CURSOR_API_KEY) {
+    return 'api_key';
+  }
+
+  // Check for credentials files
+  const credentialPaths = [
+    path.join(os.homedir(), '.cursor', 'credentials.json'),
+    path.join(os.homedir(), '.config', 'cursor', 'credentials.json'),
+    path.join(os.homedir(), '.cursor', 'auth.json'),
+    path.join(os.homedir(), '.config', 'cursor', 'auth.json'),
+  ];
+
+  for (const credPath of credentialPaths) {
+    try {
+      if (fs.existsSync(credPath)) {
+        const content = fs.readFileSync(credPath, 'utf8');
+        const creds = JSON.parse(content);
+        if (creds.accessToken || creds.token || creds.apiKey) {
+          return 'cli';
+        }
+      }
+    } catch {
+      // Invalid credentials file
+    }
+  }
+
+  // Try a simple command
+  try {
+    const result = await getCliVersion(command, ['--version'], 3000);
+    if (result) {
+      return 'cli';
+    }
+  } catch {
+    // Version check failed
+  }
+
+  return 'none';
+}
+
+/**
+ * Get installation instructions for a provider
+ */
+export function getInstallInstructions(
+  provider: keyof typeof CLI_CONFIGS,
+  platform: NodeJS.Platform = process.platform
+): string {
+  const config = CLI_CONFIGS[provider];
+  const command = config.installCommands[platform as keyof typeof config.installCommands];
+
+  if (!command) {
+    return `No installation instructions available for ${provider} on ${platform}`;
+  }
+
+  return command;
+}
+
+/**
+ * Get platform-specific CLI paths and versions
+ */
+export function getPlatformCliPaths(provider: keyof typeof CLI_CONFIGS): string[] {
+  const config = CLI_CONFIGS[provider];
+  const platform = process.platform;
+
+  switch (platform) {
+    case 'darwin':
+      return [
+        `/usr/local/bin/${config.commands[0]}`,
+        `/opt/homebrew/bin/${config.commands[0]}`,
+        path.join(os.homedir(), '.local', 'bin', config.commands[0]),
+      ];
+
+    case 'linux':
+      return [
+        `/usr/bin/${config.commands[0]}`,
+        `/usr/local/bin/${config.commands[0]}`,
+        path.join(os.homedir(), '.local', 'bin', config.commands[0]),
+        path.join(os.homedir(), '.npm', 'global', 'bin', config.commands[0]),
+      ];
+
+    case 'win32':
+      return [
+        path.join(
+          os.homedir(),
+          'AppData',
+          'Local',
+          'Programs',
+          config.commands[0],
+          `${config.commands[0]}.exe`
+        ),
+        path.join(process.env.ProgramFiles || '', config.commands[0], `${config.commands[0]}.exe`),
+        path.join(
+          process.env.ProgramFiles || '',
+          config.commands[0],
+          'bin',
+          `${config.commands[0]}.exe`
+        ),
+      ];
+
+    default:
+      return [];
+  }
+}
+
+/**
+ * Validate CLI installation
+ */
+export function validateCliInstallation(cliInfo: CliInfo): {
+  valid: boolean;
+  issues: string[];
+} {
+  const issues: string[] = [];
+
+  if (!cliInfo.installed) {
+    issues.push('CLI is not installed');
+  }
+
+  if (cliInfo.installed && !cliInfo.version) {
+    issues.push('Could not determine CLI version');
+  }
+
+  if (cliInfo.installed && cliInfo.authMethod === 'none') {
+    issues.push('CLI is not authenticated');
+  }
+
+  return {
+    valid: issues.length === 0,
+    issues,
+  };
+}

apps/server/src/lib/codex-auth.ts

@@ -0,0 +1,68 @@
+/**
+ * Shared utility for checking Codex CLI authentication status
+ *
+ * Uses 'codex login status' command to verify authentication.
+ * Never assumes authenticated - only returns true if CLI confirms.
+ */
+
+import { spawnProcess } from '@automaker/platform';
+import { findCodexCliPath } from '@automaker/platform';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('CodexAuth');
+
+const CODEX_COMMAND = 'codex';
+const OPENAI_API_KEY_ENV = 'OPENAI_API_KEY';
+
+export interface CodexAuthCheckResult {
+  authenticated: boolean;
+  method: 'api_key_env' | 'cli_authenticated' | 'none';
+}
+
+/**
+ * Check Codex authentication status using 'codex login status' command
+ *
+ * @param cliPath Optional CLI path. If not provided, will attempt to find it.
+ * @returns Authentication status and method
+ */
+export async function checkCodexAuthentication(
+  cliPath?: string | null
+): Promise<CodexAuthCheckResult> {
+  const resolvedCliPath = cliPath || (await findCodexCliPath());
+  const hasApiKey = !!process.env[OPENAI_API_KEY_ENV];
+
+  // If CLI is not installed, cannot be authenticated
+  if (!resolvedCliPath) {
+    logger.info('CLI not found');
+    return { authenticated: false, method: 'none' };
+  }
+
+  try {
+    const result = await spawnProcess({
+      command: resolvedCliPath || CODEX_COMMAND,
+      args: ['login', 'status'],
+      cwd: process.cwd(),
+      env: {
+        ...process.env,
+        TERM: 'dumb', // Avoid interactive output
+      },
+    });
+
+    // Check both stdout and stderr for "logged in" - Codex CLI outputs to stderr
+    const combinedOutput = (result.stdout + result.stderr).toLowerCase();
+    const isLoggedIn = combinedOutput.includes('logged in');
+
+    if (result.exitCode === 0 && isLoggedIn) {
+      // Determine auth method based on what we know
+      const method = hasApiKey ? 'api_key_env' : 'cli_authenticated';
+      logger.info(`✓ Authenticated (${method})`);
+      return { authenticated: true, method };
+    }
+
+    logger.info('Not authenticated');
+    return { authenticated: false, method: 'none' };
+  } catch (error) {
+    logger.error('Failed to check authentication:', error);
+    return { authenticated: false, method: 'none' };
+  }
+}

apps/server/src/lib/error-handler.ts

@@ -0,0 +1,414 @@
+/**
+ * Unified Error Handling System for CLI Providers
+ *
+ * Provides consistent error classification, user-friendly messages, and debugging support
+ * across all AI providers (Claude, Codex, Cursor)
+ */
+
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('ErrorHandler');
+
+export enum ErrorType {
+  AUTHENTICATION = 'authentication',
+  BILLING = 'billing',
+  RATE_LIMIT = 'rate_limit',
+  NETWORK = 'network',
+  TIMEOUT = 'timeout',
+  VALIDATION = 'validation',
+  PERMISSION = 'permission',
+  CLI_NOT_FOUND = 'cli_not_found',
+  CLI_NOT_INSTALLED = 'cli_not_installed',
+  MODEL_NOT_SUPPORTED = 'model_not_supported',
+  INVALID_REQUEST = 'invalid_request',
+  SERVER_ERROR = 'server_error',
+  UNKNOWN = 'unknown',
+}
+
+export enum ErrorSeverity {
+  LOW = 'low',
+  MEDIUM = 'medium',
+  HIGH = 'high',
+  CRITICAL = 'critical',
+}
+
+export interface ErrorClassification {
+  type: ErrorType;
+  severity: ErrorSeverity;
+  userMessage: string;
+  technicalMessage: string;
+  suggestedAction?: string;
+  retryable: boolean;
+  provider?: string;
+  context?: Record<string, any>;
+}
+
+export interface ErrorPattern {
+  type: ErrorType;
+  severity: ErrorSeverity;
+  patterns: RegExp[];
+  userMessage: string;
+  suggestedAction?: string;
+  retryable: boolean;
+}
+
+/**
+ * Error patterns for different types of errors
+ */
+const ERROR_PATTERNS: ErrorPattern[] = [
+  // Authentication errors
+  {
+    type: ErrorType.AUTHENTICATION,
+    severity: ErrorSeverity.HIGH,
+    patterns: [
+      /unauthorized/i,
+      /authentication.*fail/i,
+      /invalid_api_key/i,
+      /invalid api key/i,
+      /not authenticated/i,
+      /please.*log/i,
+      /token.*revoked/i,
+      /oauth.*error/i,
+      /credentials.*invalid/i,
+    ],
+    userMessage: 'Authentication failed. Please check your API key or login credentials.',
+    suggestedAction:
+      "Verify your API key is correct and hasn't expired, or run the CLI login command.",
+    retryable: false,
+  },
+
+  // Billing errors
+  {
+    type: ErrorType.BILLING,
+    severity: ErrorSeverity.HIGH,
+    patterns: [
+      /credit.*balance.*low/i,
+      /insufficient.*credit/i,
+      /billing.*issue/i,
+      /payment.*required/i,
+      /usage.*exceeded/i,
+      /quota.*exceeded/i,
+      /add.*credit/i,
+    ],
+    userMessage: 'Account has insufficient credits or billing issues.',
+    suggestedAction: 'Please add credits to your account or check your billing settings.',
+    retryable: false,
+  },
+
+  // Rate limit errors
+  {
+    type: ErrorType.RATE_LIMIT,
+    severity: ErrorSeverity.MEDIUM,
+    patterns: [
+      /rate.*limit/i,
+      /too.*many.*request/i,
+      /limit.*reached/i,
+      /try.*later/i,
+      /429/i,
+      /reset.*time/i,
+      /upgrade.*plan/i,
+    ],
+    userMessage: 'Rate limit reached. Please wait before trying again.',
+    suggestedAction: 'Wait a few minutes before retrying, or consider upgrading your plan.',
+    retryable: true,
+  },
+
+  // Network errors
+  {
+    type: ErrorType.NETWORK,
+    severity: ErrorSeverity.MEDIUM,
+    patterns: [/network/i, /connection/i, /dns/i, /timeout/i, /econnrefused/i, /enotfound/i],
+    userMessage: 'Network connection issue.',
+    suggestedAction: 'Check your internet connection and try again.',
+    retryable: true,
+  },
+
+  // Timeout errors
+  {
+    type: ErrorType.TIMEOUT,
+    severity: ErrorSeverity.MEDIUM,
+    patterns: [/timeout/i, /aborted/i, /time.*out/i],
+    userMessage: 'Operation timed out.',
+    suggestedAction: 'Try again with a simpler request or check your connection.',
+    retryable: true,
+  },
+
+  // Permission errors
+  {
+    type: ErrorType.PERMISSION,
+    severity: ErrorSeverity.HIGH,
+    patterns: [/permission.*denied/i, /access.*denied/i, /forbidden/i, /403/i, /not.*authorized/i],
+    userMessage: 'Permission denied.',
+    suggestedAction: 'Check if you have the required permissions for this operation.',
+    retryable: false,
+  },
+
+  // CLI not found
+  {
+    type: ErrorType.CLI_NOT_FOUND,
+    severity: ErrorSeverity.HIGH,
+    patterns: [/command not found/i, /not recognized/i, /not.*installed/i, /ENOENT/i],
+    userMessage: 'CLI tool not found.',
+    suggestedAction: "Please install the required CLI tool and ensure it's in your PATH.",
+    retryable: false,
+  },
+
+  // Model not supported
+  {
+    type: ErrorType.MODEL_NOT_SUPPORTED,
+    severity: ErrorSeverity.HIGH,
+    patterns: [/model.*not.*support/i, /unknown.*model/i, /invalid.*model/i],
+    userMessage: 'Model not supported.',
+    suggestedAction: 'Check available models and use a supported one.',
+    retryable: false,
+  },
+
+  // Server errors
+  {
+    type: ErrorType.SERVER_ERROR,
+    severity: ErrorSeverity.HIGH,
+    patterns: [/internal.*server/i, /server.*error/i, /500/i, /502/i, /503/i, /504/i],
+    userMessage: 'Server error occurred.',
+    suggestedAction: 'Try again in a few minutes or contact support if the issue persists.',
+    retryable: true,
+  },
+];
+
+/**
+ * Classify an error into a specific type with user-friendly message
+ */
+export function classifyError(
+  error: unknown,
+  provider?: string,
+  context?: Record<string, any>
+): ErrorClassification {
+  const errorText = getErrorText(error);
+
+  // Try to match against known patterns
+  for (const pattern of ERROR_PATTERNS) {
+    for (const regex of pattern.patterns) {
+      if (regex.test(errorText)) {
+        return {
+          type: pattern.type,
+          severity: pattern.severity,
+          userMessage: pattern.userMessage,
+          technicalMessage: errorText,
+          suggestedAction: pattern.suggestedAction,
+          retryable: pattern.retryable,
+          provider,
+          context,
+        };
+      }
+    }
+  }
+
+  // Unknown error
+  return {
+    type: ErrorType.UNKNOWN,
+    severity: ErrorSeverity.MEDIUM,
+    userMessage: 'An unexpected error occurred.',
+    technicalMessage: errorText,
+    suggestedAction: 'Please try again or contact support if the issue persists.',
+    retryable: true,
+    provider,
+    context,
+  };
+}
+
+/**
+ * Get a user-friendly error message
+ */
+export function getUserFriendlyErrorMessage(error: unknown, provider?: string): string {
+  const classification = classifyError(error, provider);
+
+  let message = classification.userMessage;
+
+  if (classification.suggestedAction) {
+    message += ` ${classification.suggestedAction}`;
+  }
+
+  // Add provider-specific context if available
+  if (provider) {
+    message = `[${provider.toUpperCase()}] ${message}`;
+  }
+
+  return message;
+}
+
+/**
+ * Check if an error is retryable
+ */
+export function isRetryableError(error: unknown): boolean {
+  const classification = classifyError(error);
+  return classification.retryable;
+}
+
+/**
+ * Check if an error is authentication-related
+ */
+export function isAuthenticationError(error: unknown): boolean {
+  const classification = classifyError(error);
+  return classification.type === ErrorType.AUTHENTICATION;
+}
+
+/**
+ * Check if an error is billing-related
+ */
+export function isBillingError(error: unknown): boolean {
+  const classification = classifyError(error);
+  return classification.type === ErrorType.BILLING;
+}
+
+/**
+ * Check if an error is rate limit related
+ */
+export function isRateLimitError(error: unknown): boolean {
+  const classification = classifyError(error);
+  return classification.type === ErrorType.RATE_LIMIT;
+}
+
+/**
+ * Get error text from various error types
+ */
+function getErrorText(error: unknown): string {
+  if (typeof error === 'string') {
+    return error;
+  }
+
+  if (error instanceof Error) {
+    return error.message;
+  }
+
+  if (typeof error === 'object' && error !== null) {
+    // Handle structured error objects
+    const errorObj = error as any;
+
+    if (errorObj.message) {
+      return errorObj.message;
+    }
+
+    if (errorObj.error?.message) {
+      return errorObj.error.message;
+    }
+
+    if (errorObj.error) {
+      return typeof errorObj.error === 'string' ? errorObj.error : JSON.stringify(errorObj.error);
+    }
+
+    return JSON.stringify(error);
+  }
+
+  return String(error);
+}
+
+/**
+ * Create a standardized error response
+ */
+export function createErrorResponse(
+  error: unknown,
+  provider?: string,
+  context?: Record<string, any>
+): {
+  success: false;
+  error: string;
+  errorType: ErrorType;
+  severity: ErrorSeverity;
+  retryable: boolean;
+  suggestedAction?: string;
+} {
+  const classification = classifyError(error, provider, context);
+
+  return {
+    success: false,
+    error: classification.userMessage,
+    errorType: classification.type,
+    severity: classification.severity,
+    retryable: classification.retryable,
+    suggestedAction: classification.suggestedAction,
+  };
+}
+
+/**
+ * Log error with full context
+ */
+export function logError(
+  error: unknown,
+  provider?: string,
+  operation?: string,
+  additionalContext?: Record<string, any>
+): void {
+  const classification = classifyError(error, provider, {
+    operation,
+    ...additionalContext,
+  });
+
+  logger.error(`Error in ${provider || 'unknown'}${operation ? ` during ${operation}` : ''}`, {
+    type: classification.type,
+    severity: classification.severity,
+    message: classification.userMessage,
+    technicalMessage: classification.technicalMessage,
+    retryable: classification.retryable,
+    suggestedAction: classification.suggestedAction,
+    context: classification.context,
+  });
+}
+
+/**
+ * Provider-specific error handlers
+ */
+export const ProviderErrorHandler = {
+  claude: {
+    classify: (error: unknown) => classifyError(error, 'claude'),
+    getUserMessage: (error: unknown) => getUserFriendlyErrorMessage(error, 'claude'),
+    isAuth: (error: unknown) => isAuthenticationError(error),
+    isBilling: (error: unknown) => isBillingError(error),
+    isRateLimit: (error: unknown) => isRateLimitError(error),
+  },
+
+  codex: {
+    classify: (error: unknown) => classifyError(error, 'codex'),
+    getUserMessage: (error: unknown) => getUserFriendlyErrorMessage(error, 'codex'),
+    isAuth: (error: unknown) => isAuthenticationError(error),
+    isBilling: (error: unknown) => isBillingError(error),
+    isRateLimit: (error: unknown) => isRateLimitError(error),
+  },
+
+  cursor: {
+    classify: (error: unknown) => classifyError(error, 'cursor'),
+    getUserMessage: (error: unknown) => getUserFriendlyErrorMessage(error, 'cursor'),
+    isAuth: (error: unknown) => isAuthenticationError(error),
+    isBilling: (error: unknown) => isBillingError(error),
+    isRateLimit: (error: unknown) => isRateLimitError(error),
+  },
+};
+
+/**
+ * Create a retry handler for retryable errors
+ */
+export function createRetryHandler(maxRetries: number = 3, baseDelay: number = 1000) {
+  return async function <T>(
+    operation: () => Promise<T>,
+    shouldRetry: (error: unknown) => boolean = isRetryableError
+  ): Promise<T> {
+    let lastError: unknown;
+
+    for (let attempt = 0; attempt <= maxRetries; attempt++) {
+      try {
+        return await operation();
+      } catch (error) {
+        lastError = error;
+
+        if (attempt === maxRetries || !shouldRetry(error)) {
+          throw error;
+        }
+
+        // Exponential backoff with jitter
+        const delay = baseDelay * Math.pow(2, attempt) + Math.random() * 1000;
+        logger.debug(`Retrying operation in ${delay}ms (attempt ${attempt + 1}/${maxRetries})`);
+        await new Promise((resolve) => setTimeout(resolve, delay));
+      }
+    }
+
+    throw lastError;
+  };
+}

apps/server/src/lib/events.ts

@@ -3,6 +3,9 @@
  */
 
 import type { EventType, EventCallback } from '@automaker/types';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('Events');
 
 // Re-export event types from shared package
 export type { EventType, EventCallback };
@@ -21,7 +24,7 @@ export function createEventEmitter(): EventEmitter {
         try {
           callback(type, payload);
         } catch (error) {
-          console.error('Error in event subscriber:', error);
+          logger.error('Error in event subscriber:', error);
         }
       }
     },

apps/server/src/lib/json-extractor.ts

@@ -0,0 +1,211 @@
+/**
+ * JSON Extraction Utilities
+ *
+ * Robust JSON extraction from AI responses that may contain markdown,
+ * code blocks, or other text mixed with JSON content.
+ *
+ * Used by various routes that parse structured output from Cursor or
+ * Claude responses when structured output is not available.
+ */
+
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('JsonExtractor');
+
+/**
+ * Logger interface for optional custom logging
+ */
+export interface JsonExtractorLogger {
+  debug: (message: string, ...args: unknown[]) => void;
+  warn?: (message: string, ...args: unknown[]) => void;
+}
+
+/**
+ * Options for JSON extraction
+ */
+export interface ExtractJsonOptions {
+  /** Custom logger (defaults to internal logger) */
+  logger?: JsonExtractorLogger;
+  /** Required key that must be present in the extracted JSON */
+  requiredKey?: string;
+  /** Whether the required key's value must be an array */
+  requireArray?: boolean;
+}
+
+/**
+ * Extract JSON from response text using multiple strategies.
+ *
+ * Strategies tried in order:
+ * 1. JSON in ```json code block
+ * 2. JSON in ``` code block (no language)
+ * 3. Find JSON object by matching braces (starting with requiredKey if specified)
+ * 4. Find any JSON object by matching braces
+ * 5. Parse entire response as JSON
+ *
+ * @param responseText - The raw response text that may contain JSON
+ * @param options - Optional extraction options
+ * @returns Parsed JSON object or null if extraction fails
+ */
+export function extractJson<T = Record<string, unknown>>(
+  responseText: string,
+  options: ExtractJsonOptions = {}
+): T | null {
+  const log = options.logger || logger;
+  const requiredKey = options.requiredKey;
+  const requireArray = options.requireArray ?? false;
+
+  /**
+   * Validate that the result has the required key/structure
+   */
+  const validateResult = (result: unknown): result is T => {
+    if (!result || typeof result !== 'object') return false;
+    if (requiredKey) {
+      const obj = result as Record<string, unknown>;
+      if (!(requiredKey in obj)) return false;
+      if (requireArray && !Array.isArray(obj[requiredKey])) return false;
+    }
+    return true;
+  };
+
+  /**
+   * Find matching closing brace by counting brackets
+   */
+  const findMatchingBrace = (text: string, startIdx: number): number => {
+    let depth = 0;
+    for (let i = startIdx; i < text.length; i++) {
+      if (text[i] === '{') depth++;
+      if (text[i] === '}') {
+        depth--;
+        if (depth === 0) {
+          return i + 1;
+        }
+      }
+    }
+    return -1;
+  };
+
+  const strategies = [
+    // Strategy 1: JSON in ```json code block
+    () => {
+      const match = responseText.match(/```json\s*([\s\S]*?)```/);
+      if (match) {
+        log.debug('Extracting JSON from ```json code block');
+        return JSON.parse(match[1].trim());
+      }
+      return null;
+    },
+
+    // Strategy 2: JSON in ``` code block (no language specified)
+    () => {
+      const match = responseText.match(/```\s*([\s\S]*?)```/);
+      if (match) {
+        const content = match[1].trim();
+        // Only try if it looks like JSON (starts with { or [)
+        if (content.startsWith('{') || content.startsWith('[')) {
+          log.debug('Extracting JSON from ``` code block');
+          return JSON.parse(content);
+        }
+      }
+      return null;
+    },
+
+    // Strategy 3: Find JSON object containing the required key (if specified)
+    () => {
+      if (!requiredKey) return null;
+
+      const searchPattern = `{"${requiredKey}"`;
+      const startIdx = responseText.indexOf(searchPattern);
+      if (startIdx === -1) return null;
+
+      const endIdx = findMatchingBrace(responseText, startIdx);
+      if (endIdx > startIdx) {
+        log.debug(`Extracting JSON with required key "${requiredKey}"`);
+        return JSON.parse(responseText.slice(startIdx, endIdx));
+      }
+      return null;
+    },
+
+    // Strategy 4: Find any JSON object by matching braces
+    () => {
+      const startIdx = responseText.indexOf('{');
+      if (startIdx === -1) return null;
+
+      const endIdx = findMatchingBrace(responseText, startIdx);
+      if (endIdx > startIdx) {
+        log.debug('Extracting JSON by brace matching');
+        return JSON.parse(responseText.slice(startIdx, endIdx));
+      }
+      return null;
+    },
+
+    // Strategy 5: Find JSON using first { to last } (may be less accurate)
+    () => {
+      const firstBrace = responseText.indexOf('{');
+      const lastBrace = responseText.lastIndexOf('}');
+      if (firstBrace !== -1 && lastBrace > firstBrace) {
+        log.debug('Extracting JSON from first { to last }');
+        return JSON.parse(responseText.slice(firstBrace, lastBrace + 1));
+      }
+      return null;
+    },
+
+    // Strategy 6: Try parsing the entire response as JSON
+    () => {
+      const trimmed = responseText.trim();
+      if (trimmed.startsWith('{') || trimmed.startsWith('[')) {
+        log.debug('Parsing entire response as JSON');
+        return JSON.parse(trimmed);
+      }
+      return null;
+    },
+  ];
+
+  for (const strategy of strategies) {
+    try {
+      const result = strategy();
+      if (validateResult(result)) {
+        log.debug('Successfully extracted JSON');
+        return result as T;
+      }
+    } catch {
+      // Strategy failed, try next
+    }
+  }
+
+  log.debug('Failed to extract JSON from response');
+  return null;
+}
+
+/**
+ * Extract JSON with a specific required key.
+ * Convenience wrapper around extractJson.
+ *
+ * @param responseText - The raw response text
+ * @param requiredKey - Key that must be present in the extracted JSON
+ * @param options - Additional options
+ * @returns Parsed JSON object or null
+ */
+export function extractJsonWithKey<T = Record<string, unknown>>(
+  responseText: string,
+  requiredKey: string,
+  options: Omit<ExtractJsonOptions, 'requiredKey'> = {}
+): T | null {
+  return extractJson<T>(responseText, { ...options, requiredKey });
+}
+
+/**
+ * Extract JSON that has a required array property.
+ * Useful for extracting responses like { "suggestions": [...] }
+ *
+ * @param responseText - The raw response text
+ * @param arrayKey - Key that must contain an array
+ * @param options - Additional options
+ * @returns Parsed JSON object or null
+ */
+export function extractJsonWithArray<T = Record<string, unknown>>(
+  responseText: string,
+  arrayKey: string,
+  options: Omit<ExtractJsonOptions, 'requiredKey' | 'requireArray'> = {}
+): T | null {
+  return extractJson<T>(responseText, { ...options, requiredKey: arrayKey, requireArray: true });
+}

apps/server/src/lib/permission-enforcer.ts

@@ -0,0 +1,173 @@
+/**
+ * Permission enforcement utilities for Cursor provider
+ */
+
+import type { CursorCliConfigFile } from '@automaker/types';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('PermissionEnforcer');
+
+export interface PermissionCheckResult {
+  allowed: boolean;
+  reason?: string;
+}
+
+/**
+ * Check if a tool call is allowed based on permissions
+ */
+export function checkToolCallPermission(
+  toolCall: any,
+  permissions: CursorCliConfigFile | null
+): PermissionCheckResult {
+  if (!permissions || !permissions.permissions) {
+    // If no permissions are configured, allow everything (backward compatibility)
+    return { allowed: true };
+  }
+
+  const { allow = [], deny = [] } = permissions.permissions;
+
+  // Check shell tool calls
+  if (toolCall.shellToolCall?.args?.command) {
+    const command = toolCall.shellToolCall.args.command;
+    const toolName = `Shell(${extractCommandName(command)})`;
+
+    // Check deny list first (deny takes precedence)
+    for (const denyRule of deny) {
+      if (matchesRule(toolName, denyRule)) {
+        return {
+          allowed: false,
+          reason: `Operation blocked by permission rule: ${denyRule}`,
+        };
+      }
+    }
+
+    // Then check allow list
+    for (const allowRule of allow) {
+      if (matchesRule(toolName, allowRule)) {
+        return { allowed: true };
+      }
+    }
+
+    return {
+      allowed: false,
+      reason: `Operation not in allow list: ${toolName}`,
+    };
+  }
+
+  // Check read tool calls
+  if (toolCall.readToolCall?.args?.path) {
+    const path = toolCall.readToolCall.args.path;
+    const toolName = `Read(${path})`;
+
+    // Check deny list first
+    for (const denyRule of deny) {
+      if (matchesRule(toolName, denyRule)) {
+        return {
+          allowed: false,
+          reason: `Read operation blocked by permission rule: ${denyRule}`,
+        };
+      }
+    }
+
+    // Then check allow list
+    for (const allowRule of allow) {
+      if (matchesRule(toolName, allowRule)) {
+        return { allowed: true };
+      }
+    }
+
+    return {
+      allowed: false,
+      reason: `Read operation not in allow list: ${toolName}`,
+    };
+  }
+
+  // Check write tool calls
+  if (toolCall.writeToolCall?.args?.path) {
+    const path = toolCall.writeToolCall.args.path;
+    const toolName = `Write(${path})`;
+
+    // Check deny list first
+    for (const denyRule of deny) {
+      if (matchesRule(toolName, denyRule)) {
+        return {
+          allowed: false,
+          reason: `Write operation blocked by permission rule: ${denyRule}`,
+        };
+      }
+    }
+
+    // Then check allow list
+    for (const allowRule of allow) {
+      if (matchesRule(toolName, allowRule)) {
+        return { allowed: true };
+      }
+    }
+
+    return {
+      allowed: false,
+      reason: `Write operation not in allow list: ${toolName}`,
+    };
+  }
+
+  // For other tool types, allow by default for now
+  return { allowed: true };
+}
+
+/**
+ * Extract the base command name from a shell command
+ */
+function extractCommandName(command: string): string {
+  // Remove leading spaces and get the first word
+  const trimmed = command.trim();
+  const firstWord = trimmed.split(/\s+/)[0];
+  return firstWord || 'unknown';
+}
+
+/**
+ * Check if a tool name matches a permission rule
+ */
+function matchesRule(toolName: string, rule: string): boolean {
+  // Exact match
+  if (toolName === rule) {
+    return true;
+  }
+
+  // Wildcard patterns
+  if (rule.includes('*')) {
+    const regex = new RegExp(rule.replace(/\*/g, '.*'));
+    return regex.test(toolName);
+  }
+
+  // Prefix match for shell commands (e.g., "Shell(git)" matches "Shell(git status)")
+  if (rule.startsWith('Shell(') && toolName.startsWith('Shell(')) {
+    const ruleCommand = rule.slice(6, -1); // Remove "Shell(" and ")"
+    const toolCommand = extractCommandName(toolName.slice(6, -1)); // Remove "Shell(" and ")"
+    return toolCommand.startsWith(ruleCommand);
+  }
+
+  return false;
+}
+
+/**
+ * Log permission violations
+ */
+export function logPermissionViolation(toolCall: any, reason: string, sessionId?: string): void {
+  const sessionIdStr = sessionId ? ` [${sessionId}]` : '';
+
+  if (toolCall.shellToolCall?.args?.command) {
+    logger.warn(
+      `Permission violation${sessionIdStr}: Shell command blocked - ${toolCall.shellToolCall.args.command} (${reason})`
+    );
+  } else if (toolCall.readToolCall?.args?.path) {
+    logger.warn(
+      `Permission violation${sessionIdStr}: Read operation blocked - ${toolCall.readToolCall.args.path} (${reason})`
+    );
+  } else if (toolCall.writeToolCall?.args?.path) {
+    logger.warn(
+      `Permission violation${sessionIdStr}: Write operation blocked - ${toolCall.writeToolCall.args.path} (${reason})`
+    );
+  } else {
+    logger.warn(`Permission violation${sessionIdStr}: Tool call blocked (${reason})`, { toolCall });
+  }
+}

apps/server/src/lib/sdk-options.ts

@@ -16,12 +16,82 @@
  */
 
 import type { Options } from '@anthropic-ai/claude-agent-sdk';
-import os from 'os';
 import path from 'path';
 import { resolveModelString } from '@automaker/model-resolver';
-import { DEFAULT_MODELS, CLAUDE_MODEL_MAP, type McpServerConfig } from '@automaker/types';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('SdkOptions');
+import {
+  DEFAULT_MODELS,
+  CLAUDE_MODEL_MAP,
+  type McpServerConfig,
+  type ThinkingLevel,
+  getThinkingTokenBudget,
+} from '@automaker/types';
 import { isPathAllowed, PathNotAllowedError, getAllowedRootDirectory } from '@automaker/platform';
 
+/**
+ * Result of sandbox compatibility check
+ */
+export interface SandboxCompatibilityResult {
+  /** Whether sandbox mode can be enabled for this path */
+  enabled: boolean;
+  /** Optional message explaining why sandbox is disabled */
+  message?: string;
+}
+
+/**
+ * Check if a working directory is compatible with sandbox mode.
+ * Some paths (like cloud storage mounts) may not work with sandboxed execution.
+ *
+ * @param cwd - The working directory to check
+ * @param sandboxRequested - Whether sandbox mode was requested by settings
+ * @returns Object indicating if sandbox can be enabled and why not if disabled
+ */
+export function checkSandboxCompatibility(
+  cwd: string,
+  sandboxRequested: boolean
+): SandboxCompatibilityResult {
+  if (!sandboxRequested) {
+    return { enabled: false };
+  }
+
+  const resolvedCwd = path.resolve(cwd);
+
+  // Check for cloud storage paths that may not be compatible with sandbox
+  const cloudStoragePatterns = [
+    // macOS mounted volumes
+    /^\/Volumes\/GoogleDrive/i,
+    /^\/Volumes\/Dropbox/i,
+    /^\/Volumes\/OneDrive/i,
+    /^\/Volumes\/iCloud/i,
+    // macOS home directory
+    /^\/Users\/[^/]+\/Google Drive/i,
+    /^\/Users\/[^/]+\/Dropbox/i,
+    /^\/Users\/[^/]+\/OneDrive/i,
+    /^\/Users\/[^/]+\/Library\/Mobile Documents/i, // iCloud
+    // Linux home directory
+    /^\/home\/[^/]+\/Google Drive/i,
+    /^\/home\/[^/]+\/Dropbox/i,
+    /^\/home\/[^/]+\/OneDrive/i,
+    // Windows
+    /^C:\\Users\\[^\\]+\\Google Drive/i,
+    /^C:\\Users\\[^\\]+\\Dropbox/i,
+    /^C:\\Users\\[^\\]+\\OneDrive/i,
+  ];
+
+  for (const pattern of cloudStoragePatterns) {
+    if (pattern.test(resolvedCwd)) {
+      return {
+        enabled: false,
+        message: `Sandbox disabled: Cloud storage path detected (${resolvedCwd}). Sandbox mode may not work correctly with cloud-synced directories.`,
+      };
+    }
+  }
+
+  return { enabled: true };
+}
+
 /**
  * Validate that a working directory is allowed by ALLOWED_ROOT_DIRECTORY.
  * This is the centralized security check for ALL AI model invocations.
@@ -48,128 +118,6 @@ export function validateWorkingDirectory(cwd: string): void {
   }
 }
 
-/**
- * Known cloud storage path patterns where sandbox mode is incompatible.
- *
- * The Claude CLI sandbox feature uses filesystem isolation that conflicts with
- * cloud storage providers' virtual filesystem implementations. This causes the
- * Claude process to exit with code 1 when sandbox is enabled for these paths.
- *
- * Affected providers (macOS paths):
- * - Dropbox: ~/Library/CloudStorage/Dropbox-*
- * - Google Drive: ~/Library/CloudStorage/GoogleDrive-*
- * - OneDrive: ~/Library/CloudStorage/OneDrive-*
- * - iCloud Drive: ~/Library/Mobile Documents/
- * - Box: ~/Library/CloudStorage/Box-*
- *
- * @see https://github.com/anthropics/claude-code/issues/XXX (TODO: file upstream issue)
- */
-
-/**
- * macOS-specific cloud storage patterns that appear under ~/Library/
- * These are specific enough to use with includes() safely.
- */
-const MACOS_CLOUD_STORAGE_PATTERNS = [
-  '/Library/CloudStorage/', // Dropbox, Google Drive, OneDrive, Box on macOS
-  '/Library/Mobile Documents/', // iCloud Drive on macOS
-] as const;
-
-/**
- * Generic cloud storage folder names that need to be anchored to the home directory
- * to avoid false positives (e.g., /home/user/my-project-about-dropbox/).
- */
-const HOME_ANCHORED_CLOUD_FOLDERS = [
-  'Google Drive', // Google Drive on some systems
-  'Dropbox', // Dropbox on Linux/alternative installs
-  'OneDrive', // OneDrive on Linux/alternative installs
-] as const;
-
-/**
- * Check if a path is within a cloud storage location.
- *
- * Cloud storage providers use virtual filesystem implementations that are
- * incompatible with the Claude CLI sandbox feature, causing process crashes.
- *
- * Uses two detection strategies:
- * 1. macOS-specific patterns (under ~/Library/) - checked via includes()
- * 2. Generic folder names - anchored to home directory to avoid false positives
- *
- * @param cwd - The working directory path to check
- * @returns true if the path is in a cloud storage location
- */
-export function isCloudStoragePath(cwd: string): boolean {
-  const resolvedPath = path.resolve(cwd);
-
-  // Check macOS-specific patterns (these are specific enough to use includes)
-  if (MACOS_CLOUD_STORAGE_PATTERNS.some((pattern) => resolvedPath.includes(pattern))) {
-    return true;
-  }
-
-  // Check home-anchored patterns to avoid false positives
-  // e.g., /home/user/my-project-about-dropbox/ should NOT match
-  const home = os.homedir();
-  for (const folder of HOME_ANCHORED_CLOUD_FOLDERS) {
-    const cloudPath = path.join(home, folder);
-    // Check if resolved path starts with the cloud storage path followed by a separator
-    // This ensures we match ~/Dropbox/project but not ~/Dropbox-archive or ~/my-dropbox-tool
-    if (resolvedPath === cloudPath || resolvedPath.startsWith(cloudPath + path.sep)) {
-      return true;
-    }
-  }
-
-  return false;
-}
-
-/**
- * Result of sandbox compatibility check
- */
-export interface SandboxCheckResult {
-  /** Whether sandbox should be enabled */
-  enabled: boolean;
-  /** If disabled, the reason why */
-  disabledReason?: 'cloud_storage' | 'user_setting';
-  /** Human-readable message for logging/UI */
-  message?: string;
-}
-
-/**
- * Determine if sandbox mode should be enabled for a given configuration.
- *
- * Sandbox mode is automatically disabled for cloud storage paths because the
- * Claude CLI sandbox feature is incompatible with virtual filesystem
- * implementations used by cloud storage providers (Dropbox, Google Drive, etc.).
- *
- * @param cwd - The working directory
- * @param enableSandboxMode - User's sandbox mode setting
- * @returns SandboxCheckResult with enabled status and reason if disabled
- */
-export function checkSandboxCompatibility(
-  cwd: string,
-  enableSandboxMode?: boolean
-): SandboxCheckResult {
-  // User has explicitly disabled sandbox mode
-  if (enableSandboxMode === false) {
-    return {
-      enabled: false,
-      disabledReason: 'user_setting',
-    };
-  }
-
-  // Check for cloud storage incompatibility (applies when enabled or undefined)
-  if (isCloudStoragePath(cwd)) {
-    return {
-      enabled: false,
-      disabledReason: 'cloud_storage',
-      message: `Sandbox mode auto-disabled: Project is in a cloud storage location (${cwd}). The Claude CLI sandbox feature is incompatible with cloud storage filesystems. To use sandbox mode, move your project to a local directory.`,
-    };
-  }
-
-  // Sandbox is compatible and enabled (true or undefined defaults to enabled)
-  return {
-    enabled: true,
-  };
-}
-
 /**
  * Tool presets for different use cases
  */
@@ -252,60 +200,51 @@ export function getModelForUseCase(
 
 /**
  * Base options that apply to all SDK calls
+ * AUTONOMOUS MODE: Always bypass permissions for fully autonomous operation
  */
 function getBaseOptions(): Partial<Options> {
   return {
-    permissionMode: 'acceptEdits',
+    permissionMode: 'bypassPermissions',
+    allowDangerouslySkipPermissions: true,
   };
 }
 
 /**
- * MCP permission options result
+ * MCP options result
  */
-interface McpPermissionOptions {
-  /** Whether tools should be restricted to a preset */
-  shouldRestrictTools: boolean;
-  /** Options to spread when MCP bypass is enabled */
-  bypassOptions: Partial<Options>;
+interface McpOptions {
   /** Options to spread for MCP servers */
   mcpServerOptions: Partial<Options>;
 }
 
 /**
  * Build MCP-related options based on configuration.
- * Centralizes the logic for determining permission modes and tool restrictions
- * when MCP servers are configured.
  *
  * @param config - The SDK options config
- * @returns Object with MCP permission settings to spread into final options
+ * @returns Object with MCP server settings to spread into final options
  */
-function buildMcpOptions(config: CreateSdkOptionsConfig): McpPermissionOptions {
-  const hasMcpServers = config.mcpServers && Object.keys(config.mcpServers).length > 0;
-  // Default to true for autonomous workflow. Security is enforced when adding servers
-  // via the security warning dialog that explains the risks.
-  const mcpAutoApprove = config.mcpAutoApproveTools ?? true;
-  const mcpUnrestricted = config.mcpUnrestrictedTools ?? true;
-
-  // Determine if we should bypass permissions based on settings
-  const shouldBypassPermissions = hasMcpServers && mcpAutoApprove;
-  // Determine if we should restrict tools (only when no MCP or unrestricted is disabled)
-  const shouldRestrictTools = !hasMcpServers || !mcpUnrestricted;
-
+function buildMcpOptions(config: CreateSdkOptionsConfig): McpOptions {
   return {
-    shouldRestrictTools,
-    // Only include bypass options when MCP is configured and auto-approve is enabled
-    bypassOptions: shouldBypassPermissions
-      ? {
-          permissionMode: 'bypassPermissions' as const,
-          // Required flag when using bypassPermissions mode
-          allowDangerouslySkipPermissions: true,
-        }
-      : {},
     // Include MCP servers if configured
     mcpServerOptions: config.mcpServers ? { mcpServers: config.mcpServers } : {},
   };
 }
 
+/**
+ * Build thinking options for SDK configuration.
+ * Converts ThinkingLevel to maxThinkingTokens for the Claude SDK.
+ *
+ * @param thinkingLevel - The thinking level to convert
+ * @returns Object with maxThinkingTokens if thinking is enabled
+ */
+function buildThinkingOptions(thinkingLevel?: ThinkingLevel): Partial<Options> {
+  const maxThinkingTokens = getThinkingTokenBudget(thinkingLevel);
+  logger.debug(
+    `buildThinkingOptions: thinkingLevel="${thinkingLevel}" -> maxThinkingTokens=${maxThinkingTokens}`
+  );
+  return maxThinkingTokens ? { maxThinkingTokens } : {};
+}
+
 /**
  * Build system prompt configuration based on autoLoadClaudeMd setting.
  * When autoLoadClaudeMd is true:
@@ -387,17 +326,11 @@ export interface CreateSdkOptionsConfig {
   /** Enable auto-loading of CLAUDE.md files via SDK's settingSources */
   autoLoadClaudeMd?: boolean;
 
-  /** Enable sandbox mode for bash command isolation */
-  enableSandboxMode?: boolean;
-
   /** MCP servers to make available to the agent */
   mcpServers?: Record<string, McpServerConfig>;
 
-  /** Auto-approve MCP tool calls without permission prompts */
-  mcpAutoApproveTools?: boolean;
-
-  /** Allow unrestricted tools when MCP servers are enabled */
-  mcpUnrestrictedTools?: boolean;
+  /** Extended thinking level for Claude models */
+  thinkingLevel?: ThinkingLevel;
 }
 
 // Re-export MCP types from @automaker/types for convenience
@@ -424,6 +357,9 @@ export function createSpecGenerationOptions(config: CreateSdkOptionsConfig): Opt
   // Build CLAUDE.md auto-loading options if enabled
   const claudeMdOptions = buildClaudeMdOptions(config);
 
+  // Build thinking options
+  const thinkingOptions = buildThinkingOptions(config.thinkingLevel);
+
   return {
     ...getBaseOptions(),
     // Override permissionMode - spec generation only needs read-only tools
@@ -435,6 +371,7 @@ export function createSpecGenerationOptions(config: CreateSdkOptionsConfig): Opt
     cwd: config.cwd,
     allowedTools: [...TOOL_PRESETS.specGeneration],
     ...claudeMdOptions,
+    ...thinkingOptions,
     ...(config.abortController && { abortController: config.abortController }),
     ...(config.outputFormat && { outputFormat: config.outputFormat }),
   };
@@ -456,6 +393,9 @@ export function createFeatureGenerationOptions(config: CreateSdkOptionsConfig):
   // Build CLAUDE.md auto-loading options if enabled
   const claudeMdOptions = buildClaudeMdOptions(config);
 
+  // Build thinking options
+  const thinkingOptions = buildThinkingOptions(config.thinkingLevel);
+
   return {
     ...getBaseOptions(),
     // Override permissionMode - feature generation only needs read-only tools
@@ -465,6 +405,7 @@ export function createFeatureGenerationOptions(config: CreateSdkOptionsConfig):
     cwd: config.cwd,
     allowedTools: [...TOOL_PRESETS.readOnly],
     ...claudeMdOptions,
+    ...thinkingOptions,
     ...(config.abortController && { abortController: config.abortController }),
   };
 }
@@ -485,6 +426,9 @@ export function createSuggestionsOptions(config: CreateSdkOptionsConfig): Option
   // Build CLAUDE.md auto-loading options if enabled
   const claudeMdOptions = buildClaudeMdOptions(config);
 
+  // Build thinking options
+  const thinkingOptions = buildThinkingOptions(config.thinkingLevel);
+
   return {
     ...getBaseOptions(),
     model: getModelForUseCase('suggestions', config.model),
@@ -492,6 +436,7 @@ export function createSuggestionsOptions(config: CreateSdkOptionsConfig): Option
     cwd: config.cwd,
     allowedTools: [...TOOL_PRESETS.readOnly],
     ...claudeMdOptions,
+    ...thinkingOptions,
     ...(config.abortController && { abortController: config.abortController }),
     ...(config.outputFormat && { outputFormat: config.outputFormat }),
   };
@@ -504,7 +449,6 @@ export function createSuggestionsOptions(config: CreateSdkOptionsConfig): Option
  * - Full tool access for code modification
  * - Standard turns for interactive sessions
  * - Model priority: explicit model > session model > chat default
- * - Sandbox mode controlled by enableSandboxMode setting (auto-disabled for cloud storage)
  * - When autoLoadClaudeMd is true, uses preset mode and settingSources for CLAUDE.md loading
  */
 export function createChatOptions(config: CreateSdkOptionsConfig): Options {
@@ -520,25 +464,17 @@ export function createChatOptions(config: CreateSdkOptionsConfig): Options {
   // Build MCP-related options
   const mcpOptions = buildMcpOptions(config);
 
-  // Check sandbox compatibility (auto-disables for cloud storage paths)
-  const sandboxCheck = checkSandboxCompatibility(config.cwd, config.enableSandboxMode);
+  // Build thinking options
+  const thinkingOptions = buildThinkingOptions(config.thinkingLevel);
 
   return {
     ...getBaseOptions(),
     model: getModelForUseCase('chat', effectiveModel),
     maxTurns: MAX_TURNS.standard,
     cwd: config.cwd,
-    // Only restrict tools if no MCP servers configured or unrestricted is disabled
-    ...(mcpOptions.shouldRestrictTools && { allowedTools: [...TOOL_PRESETS.chat] }),
-    // Apply MCP bypass options if configured
-    ...mcpOptions.bypassOptions,
-    ...(sandboxCheck.enabled && {
-      sandbox: {
-        enabled: true,
-        autoAllowBashIfSandboxed: true,
-      },
-    }),
+    allowedTools: [...TOOL_PRESETS.chat],
     ...claudeMdOptions,
+    ...thinkingOptions,
     ...(config.abortController && { abortController: config.abortController }),
     ...mcpOptions.mcpServerOptions,
   };
@@ -551,7 +487,6 @@ export function createChatOptions(config: CreateSdkOptionsConfig): Options {
  * - Full tool access for code modification and implementation
  * - Extended turns for thorough feature implementation
  * - Uses default model (can be overridden)
- * - Sandbox mode controlled by enableSandboxMode setting (auto-disabled for cloud storage)
  * - When autoLoadClaudeMd is true, uses preset mode and settingSources for CLAUDE.md loading
  */
 export function createAutoModeOptions(config: CreateSdkOptionsConfig): Options {
@@ -564,25 +499,17 @@ export function createAutoModeOptions(config: CreateSdkOptionsConfig): Options {
   // Build MCP-related options
   const mcpOptions = buildMcpOptions(config);
 
-  // Check sandbox compatibility (auto-disables for cloud storage paths)
-  const sandboxCheck = checkSandboxCompatibility(config.cwd, config.enableSandboxMode);
+  // Build thinking options
+  const thinkingOptions = buildThinkingOptions(config.thinkingLevel);
 
   return {
     ...getBaseOptions(),
     model: getModelForUseCase('auto', config.model),
     maxTurns: MAX_TURNS.maximum,
     cwd: config.cwd,
-    // Only restrict tools if no MCP servers configured or unrestricted is disabled
-    ...(mcpOptions.shouldRestrictTools && { allowedTools: [...TOOL_PRESETS.fullAccess] }),
-    // Apply MCP bypass options if configured
-    ...mcpOptions.bypassOptions,
-    ...(sandboxCheck.enabled && {
-      sandbox: {
-        enabled: true,
-        autoAllowBashIfSandboxed: true,
-      },
-    }),
+    allowedTools: [...TOOL_PRESETS.fullAccess],
     ...claudeMdOptions,
+    ...thinkingOptions,
     ...(config.abortController && { abortController: config.abortController }),
     ...mcpOptions.mcpServerOptions,
   };
@@ -598,7 +525,6 @@ export function createCustomOptions(
   config: CreateSdkOptionsConfig & {
     maxTurns?: number;
     allowedTools?: readonly string[];
-    sandbox?: { enabled: boolean; autoAllowBashIfSandboxed?: boolean };
   }
 ): Options {
   // Validate working directory before creating options
@@ -610,23 +536,22 @@ export function createCustomOptions(
   // Build MCP-related options
   const mcpOptions = buildMcpOptions(config);
 
-  // For custom options: use explicit allowedTools if provided, otherwise use preset based on MCP settings
+  // Build thinking options
+  const thinkingOptions = buildThinkingOptions(config.thinkingLevel);
+
+  // For custom options: use explicit allowedTools if provided, otherwise default to readOnly
   const effectiveAllowedTools = config.allowedTools
     ? [...config.allowedTools]
-    : mcpOptions.shouldRestrictTools
-      ? [...TOOL_PRESETS.readOnly]
-      : undefined;
+    : [...TOOL_PRESETS.readOnly];
 
   return {
     ...getBaseOptions(),
     model: getModelForUseCase('default', config.model),
     maxTurns: config.maxTurns ?? MAX_TURNS.maximum,
     cwd: config.cwd,
-    ...(effectiveAllowedTools && { allowedTools: effectiveAllowedTools }),
-    ...(config.sandbox && { sandbox: config.sandbox }),
-    // Apply MCP bypass options if configured
-    ...mcpOptions.bypassOptions,
+    allowedTools: effectiveAllowedTools,
     ...claudeMdOptions,
+    ...thinkingOptions,
     ...(config.abortController && { abortController: config.abortController }),
     ...mcpOptions.mcpServerOptions,
   };

apps/server/src/lib/settings-helpers.ts

@@ -55,34 +55,6 @@ export async function getAutoLoadClaudeMdSetting(
   }
 }
 
-/**
- * Get the enableSandboxMode setting from global settings.
- * Returns false if settings service is not available.
- *
- * @param settingsService - Optional settings service instance
- * @param logPrefix - Prefix for log messages (e.g., '[AgentService]')
- * @returns Promise resolving to the enableSandboxMode setting value
- */
-export async function getEnableSandboxModeSetting(
-  settingsService?: SettingsService | null,
-  logPrefix = '[SettingsHelper]'
-): Promise<boolean> {
-  if (!settingsService) {
-    logger.info(`${logPrefix} SettingsService not available, sandbox mode disabled`);
-    return false;
-  }
-
-  try {
-    const globalSettings = await settingsService.getGlobalSettings();
-    const result = globalSettings.enableSandboxMode ?? false;
-    logger.info(`${logPrefix} enableSandboxMode from global settings: ${result}`);
-    return result;
-  } catch (error) {
-    logger.error(`${logPrefix} Failed to load enableSandboxMode setting:`, error);
-    throw error;
-  }
-}
-
 /**
  * Filters out CLAUDE.md from context files when autoLoadClaudeMd is enabled
  * and rebuilds the formatted prompt without it.
@@ -191,41 +163,6 @@ export async function getMCPServersFromSettings(
   }
 }
 
-/**
- * Get MCP permission settings from global settings.
- *
- * @param settingsService - Optional settings service instance
- * @param logPrefix - Prefix for log messages (e.g., '[AgentService]')
- * @returns Promise resolving to MCP permission settings
- */
-export async function getMCPPermissionSettings(
-  settingsService?: SettingsService | null,
-  logPrefix = '[SettingsHelper]'
-): Promise<{ mcpAutoApproveTools: boolean; mcpUnrestrictedTools: boolean }> {
-  // Default to true for autonomous workflow. Security is enforced when adding servers
-  // via the security warning dialog that explains the risks.
-  const defaults = { mcpAutoApproveTools: true, mcpUnrestrictedTools: true };
-
-  if (!settingsService) {
-    return defaults;
-  }
-
-  try {
-    const globalSettings = await settingsService.getGlobalSettings();
-    const result = {
-      mcpAutoApproveTools: globalSettings.mcpAutoApproveTools ?? true,
-      mcpUnrestrictedTools: globalSettings.mcpUnrestrictedTools ?? true,
-    };
-    logger.info(
-      `${logPrefix} MCP permission settings: autoApprove=${result.mcpAutoApproveTools}, unrestricted=${result.mcpUnrestrictedTools}`
-    );
-    return result;
-  } catch (error) {
-    logger.error(`${logPrefix} Failed to load MCP permission settings:`, error);
-    return defaults;
-  }
-}
-
 /**
  * Convert a settings MCPServerConfig to SDK McpServerConfig format.
  * Validates required fields and throws informative errors if missing.
@@ -304,3 +241,83 @@ export async function getPromptCustomization(
     enhancement: mergeEnhancementPrompts(customization.enhancement),
   };
 }
+
+/**
+ * Get Skills configuration from settings.
+ * Returns configuration for enabling skills and which sources to load from.
+ *
+ * @param settingsService - Settings service instance
+ * @returns Skills configuration with enabled state, sources, and tool inclusion flag
+ */
+export async function getSkillsConfiguration(settingsService: SettingsService): Promise<{
+  enabled: boolean;
+  sources: Array<'user' | 'project'>;
+  shouldIncludeInTools: boolean;
+}> {
+  const settings = await settingsService.getGlobalSettings();
+  const enabled = settings.enableSkills ?? true; // Default enabled
+  const sources = settings.skillsSources ?? ['user', 'project']; // Default both sources
+
+  return {
+    enabled,
+    sources,
+    shouldIncludeInTools: enabled && sources.length > 0,
+  };
+}
+
+/**
+ * Get Subagents configuration from settings.
+ * Returns configuration for enabling subagents and which sources to load from.
+ *
+ * @param settingsService - Settings service instance
+ * @returns Subagents configuration with enabled state, sources, and tool inclusion flag
+ */
+export async function getSubagentsConfiguration(settingsService: SettingsService): Promise<{
+  enabled: boolean;
+  sources: Array<'user' | 'project'>;
+  shouldIncludeInTools: boolean;
+}> {
+  const settings = await settingsService.getGlobalSettings();
+  const enabled = settings.enableSubagents ?? true; // Default enabled
+  const sources = settings.subagentsSources ?? ['user', 'project']; // Default both sources
+
+  return {
+    enabled,
+    sources,
+    shouldIncludeInTools: enabled && sources.length > 0,
+  };
+}
+
+/**
+ * Get custom subagents from settings, merging global and project-level definitions.
+ * Project-level subagents take precedence over global ones with the same name.
+ *
+ * @param settingsService - Settings service instance
+ * @param projectPath - Path to the project for loading project-specific subagents
+ * @returns Record of agent names to definitions, or undefined if none configured
+ */
+export async function getCustomSubagents(
+  settingsService: SettingsService,
+  projectPath?: string
+): Promise<Record<string, import('@automaker/types').AgentDefinition> | undefined> {
+  // Get global subagents
+  const globalSettings = await settingsService.getGlobalSettings();
+  const globalSubagents = globalSettings.customSubagents || {};
+
+  // If no project path, return only global subagents
+  if (!projectPath) {
+    return Object.keys(globalSubagents).length > 0 ? globalSubagents : undefined;
+  }
+
+  // Get project-specific subagents
+  const projectSettings = await settingsService.getProjectSettings(projectPath);
+  const projectSubagents = projectSettings.customSubagents || {};
+
+  // Merge: project-level takes precedence
+  const merged = {
+    ...globalSubagents,
+    ...projectSubagents,
+  };
+
+  return Object.keys(merged).length > 0 ? merged : undefined;
+}

apps/server/src/lib/version.ts

@@ -5,6 +5,9 @@
 import { readFileSync } from 'fs';
 import { fileURLToPath } from 'url';
 import { dirname, join } from 'path';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('Version');
 
 const __filename = fileURLToPath(import.meta.url);
 const __dirname = dirname(__filename);
@@ -27,7 +30,7 @@ export function getVersion(): string {
     cachedVersion = version;
     return version;
   } catch (error) {
-    console.warn('Failed to read version from package.json:', error);
+    logger.warn('Failed to read version from package.json:', error);
     return '0.0.0';
   }
 }

apps/server/src/lib/worktree-metadata.ts

@@ -21,6 +21,12 @@ export interface WorktreeMetadata {
   branch: string;
   createdAt: string;
   pr?: WorktreePRInfo;
+  /** Whether the init script has been executed for this worktree */
+  initScriptRan?: boolean;
+  /** Status of the init script execution */
+  initScriptStatus?: 'running' | 'success' | 'failed';
+  /** Error message if init script failed */
+  initScriptError?: string;
 }
 
 /**

apps/server/src/providers/claude-provider.ts

@@ -7,7 +7,10 @@
 
 import { query, type Options } from '@anthropic-ai/claude-agent-sdk';
 import { BaseProvider } from './base-provider.js';
-import { classifyError, getUserFriendlyErrorMessage } from '@automaker/utils';
+import { classifyError, getUserFriendlyErrorMessage, createLogger } from '@automaker/utils';
+
+const logger = createLogger('ClaudeProvider');
+import { getThinkingTokenBudget, validateBareModelId } from '@automaker/types';
 import type {
   ExecuteOptions,
   ProviderMessage,
@@ -50,6 +53,10 @@ export class ClaudeProvider extends BaseProvider {
    * Execute a query using Claude Agent SDK
    */
   async *executeQuery(options: ExecuteOptions): AsyncGenerator<ProviderMessage> {
+    // Validate that model doesn't have a provider prefix
+    // AgentService should strip prefixes before passing to providers
+    validateBareModelId(options.model, 'ClaudeProvider');
+
     const {
       prompt,
       model,
@@ -60,24 +67,13 @@ export class ClaudeProvider extends BaseProvider {
       abortController,
       conversationHistory,
       sdkSessionId,
+      thinkingLevel,
     } = options;
 
+    // Convert thinking level to token budget
+    const maxThinkingTokens = getThinkingTokenBudget(thinkingLevel);
+
     // Build Claude SDK options
-    // MCP permission logic - determines how to handle tool permissions when MCP servers are configured.
-    // This logic mirrors buildMcpOptions() in sdk-options.ts but is applied here since
-    // the provider is the final point where SDK options are constructed.
-    const hasMcpServers = options.mcpServers && Object.keys(options.mcpServers).length > 0;
-    // Default to true for autonomous workflow. Security is enforced when adding servers
-    // via the security warning dialog that explains the risks.
-    const mcpAutoApprove = options.mcpAutoApproveTools ?? true;
-    const mcpUnrestricted = options.mcpUnrestrictedTools ?? true;
-    const defaultTools = ['Read', 'Write', 'Edit', 'Glob', 'Grep', 'Bash', 'WebSearch', 'WebFetch'];
-
-    // Determine permission mode based on settings
-    const shouldBypassPermissions = hasMcpServers && mcpAutoApprove;
-    // Determine if we should restrict tools (only when no MCP or unrestricted is disabled)
-    const shouldRestrictTools = !hasMcpServers || !mcpUnrestricted;
-
     const sdkOptions: Options = {
       model,
       systemPrompt,
@@ -85,13 +81,11 @@ export class ClaudeProvider extends BaseProvider {
       cwd,
       // Pass only explicitly allowed environment variables to SDK
       env: buildEnv(),
-      // Only restrict tools if explicitly set OR (no MCP / unrestricted disabled)
-      ...(allowedTools && shouldRestrictTools && { allowedTools }),
-      ...(!allowedTools && shouldRestrictTools && { allowedTools: defaultTools }),
-      // When MCP servers are configured and auto-approve is enabled, use bypassPermissions
-      permissionMode: shouldBypassPermissions ? 'bypassPermissions' : 'default',
-      // Required when using bypassPermissions mode
-      ...(shouldBypassPermissions && { allowDangerouslySkipPermissions: true }),
+      // Pass through allowedTools if provided by caller (decided by sdk-options.ts)
+      ...(allowedTools && { allowedTools }),
+      // AUTONOMOUS MODE: Always bypass permissions for fully autonomous operation
+      permissionMode: 'bypassPermissions',
+      allowDangerouslySkipPermissions: true,
       abortController,
       // Resume existing SDK session if we have a session ID
       ...(sdkSessionId && conversationHistory && conversationHistory.length > 0
@@ -99,10 +93,14 @@ export class ClaudeProvider extends BaseProvider {
         : {}),
       // Forward settingSources for CLAUDE.md file loading
       ...(options.settingSources && { settingSources: options.settingSources }),
-      // Forward sandbox configuration
-      ...(options.sandbox && { sandbox: options.sandbox }),
       // Forward MCP servers configuration
       ...(options.mcpServers && { mcpServers: options.mcpServers }),
+      // Extended thinking configuration
+      ...(maxThinkingTokens && { maxThinkingTokens }),
+      // Subagents configuration for specialized task delegation
+      ...(options.agents && { agents: options.agents }),
+      // Pass through outputFormat for structured JSON outputs
+      ...(options.outputFormat && { outputFormat: options.outputFormat }),
     };
 
     // Build prompt payload
@@ -140,7 +138,7 @@ export class ClaudeProvider extends BaseProvider {
       const errorInfo = classifyError(error);
       const userMessage = getUserFriendlyErrorMessage(error);
 
-      console.error('[ClaudeProvider] executeQuery() error during execution:', {
+      logger.error('executeQuery() error during execution:', {
         type: errorInfo.type,
         message: errorInfo.message,
         isRateLimit: errorInfo.isRateLimit,

apps/server/src/providers/cli-provider.ts

@@ -0,0 +1,558 @@
+/**
+ * CliProvider - Abstract base class for CLI-based AI providers
+ *
+ * Provides common infrastructure for CLI tools that spawn subprocesses
+ * and stream JSONL output. Handles:
+ * - Platform-specific CLI detection (PATH, common locations)
+ * - Windows execution strategies (WSL, npx, direct, cmd)
+ * - JSONL subprocess spawning and streaming
+ * - Error mapping infrastructure
+ *
+ * @example
+ * ```typescript
+ * class CursorProvider extends CliProvider {
+ *   getCliName(): string { return 'cursor-agent'; }
+ *   getSpawnConfig(): CliSpawnConfig {
+ *     return {
+ *       windowsStrategy: 'wsl',
+ *       commonPaths: {
+ *         linux: ['~/.local/bin/cursor-agent'],
+ *         darwin: ['~/.local/bin/cursor-agent'],
+ *       }
+ *     };
+ *   }
+ *   // ... implement abstract methods
+ * }
+ * ```
+ */
+
+import { execSync } from 'child_process';
+import * as fs from 'fs';
+import * as path from 'path';
+import * as os from 'os';
+import { BaseProvider } from './base-provider.js';
+import type { ProviderConfig, ExecuteOptions, ProviderMessage } from './types.js';
+import {
+  spawnJSONLProcess,
+  type SubprocessOptions,
+  isWslAvailable,
+  findCliInWsl,
+  createWslCommand,
+  windowsToWslPath,
+  type WslCliResult,
+} from '@automaker/platform';
+import { createLogger, isAbortError } from '@automaker/utils';
+
+/**
+ * Spawn strategy for CLI tools on Windows
+ *
+ * Different CLI tools require different execution strategies:
+ * - 'wsl': Requires WSL, CLI only available on Linux/macOS (e.g., cursor-agent)
+ * - 'npx': Installed globally via npm/npx, use `npx <package>` to run
+ * - 'direct': Native Windows binary, can spawn directly
+ * - 'cmd': Windows batch file (.cmd/.bat), needs cmd.exe shell
+ */
+export type SpawnStrategy = 'wsl' | 'npx' | 'direct' | 'cmd';
+
+/**
+ * Configuration for CLI tool spawning
+ */
+export interface CliSpawnConfig {
+  /** How to spawn on Windows */
+  windowsStrategy: SpawnStrategy;
+
+  /** NPX package name (required if windowsStrategy is 'npx') */
+  npxPackage?: string;
+
+  /** Preferred WSL distribution (if windowsStrategy is 'wsl') */
+  wslDistribution?: string;
+
+  /**
+   * Common installation paths per platform
+   * Use ~ for home directory (will be expanded)
+   * Keys: 'linux', 'darwin', 'win32'
+   */
+  commonPaths: Record<string, string[]>;
+
+  /** Version check command (defaults to --version) */
+  versionCommand?: string;
+}
+
+/**
+ * CLI error information for consistent error handling
+ */
+export interface CliErrorInfo {
+  code: string;
+  message: string;
+  recoverable: boolean;
+  suggestion?: string;
+}
+
+/**
+ * Detection result from CLI path finding
+ */
+export interface CliDetectionResult {
+  /** Path to the CLI (or 'npx' for npx strategy) */
+  cliPath: string | null;
+  /** Whether using WSL mode */
+  useWsl: boolean;
+  /** WSL path if using WSL */
+  wslCliPath?: string;
+  /** WSL distribution if using WSL */
+  wslDistribution?: string;
+  /** Detected strategy used */
+  strategy: SpawnStrategy | 'native';
+}
+
+// Create logger for CLI operations
+const cliLogger = createLogger('CliProvider');
+
+/**
+ * Abstract base class for CLI-based providers
+ *
+ * Subclasses must implement:
+ * - getCliName(): CLI executable name
+ * - getSpawnConfig(): Platform-specific spawn configuration
+ * - buildCliArgs(): Convert ExecuteOptions to CLI arguments
+ * - normalizeEvent(): Convert CLI output to ProviderMessage
+ */
+export abstract class CliProvider extends BaseProvider {
+  // CLI detection results (cached after first detection)
+  protected cliPath: string | null = null;
+  protected useWsl: boolean = false;
+  protected wslCliPath: string | null = null;
+  protected wslDistribution: string | undefined = undefined;
+  protected detectedStrategy: SpawnStrategy | 'native' = 'native';
+
+  // NPX args (used when strategy is 'npx')
+  protected npxArgs: string[] = [];
+
+  constructor(config: ProviderConfig = {}) {
+    super(config);
+    // Detection happens lazily on first use
+  }
+
+  // ==========================================================================
+  // Abstract methods - must be implemented by subclasses
+  // ==========================================================================
+
+  /**
+   * Get the CLI executable name (e.g., 'cursor-agent', 'aider')
+   */
+  abstract getCliName(): string;
+
+  /**
+   * Get spawn configuration for this CLI
+   */
+  abstract getSpawnConfig(): CliSpawnConfig;
+
+  /**
+   * Build CLI arguments from execution options
+   * @param options Execution options
+   * @returns Array of CLI arguments
+   */
+  abstract buildCliArgs(options: ExecuteOptions): string[];
+
+  /**
+   * Normalize a raw CLI event to ProviderMessage format
+   * @param event Raw event from CLI JSONL output
+   * @returns Normalized ProviderMessage or null to skip
+   */
+  abstract normalizeEvent(event: unknown): ProviderMessage | null;
+
+  // ==========================================================================
+  // Optional overrides
+  // ==========================================================================
+
+  /**
+   * Map CLI stderr/exit code to error info
+   * Override to provide CLI-specific error mapping
+   */
+  protected mapError(stderr: string, exitCode: number | null): CliErrorInfo {
+    const lower = stderr.toLowerCase();
+
+    // Common authentication errors
+    if (
+      lower.includes('not authenticated') ||
+      lower.includes('please log in') ||
+      lower.includes('unauthorized')
+    ) {
+      return {
+        code: 'NOT_AUTHENTICATED',
+        message: `${this.getCliName()} is not authenticated`,
+        recoverable: true,
+        suggestion: `Run "${this.getCliName()} login" to authenticate`,
+      };
+    }
+
+    // Rate limiting
+    if (
+      lower.includes('rate limit') ||
+      lower.includes('too many requests') ||
+      lower.includes('429')
+    ) {
+      return {
+        code: 'RATE_LIMITED',
+        message: 'API rate limit exceeded',
+        recoverable: true,
+        suggestion: 'Wait a few minutes and try again',
+      };
+    }
+
+    // Network errors
+    if (
+      lower.includes('network') ||
+      lower.includes('connection') ||
+      lower.includes('econnrefused') ||
+      lower.includes('timeout')
+    ) {
+      return {
+        code: 'NETWORK_ERROR',
+        message: 'Network connection error',
+        recoverable: true,
+        suggestion: 'Check your internet connection and try again',
+      };
+    }
+
+    // Process killed
+    if (exitCode === 137 || lower.includes('killed') || lower.includes('sigterm')) {
+      return {
+        code: 'PROCESS_CRASHED',
+        message: 'Process was terminated',
+        recoverable: true,
+        suggestion: 'The process may have run out of memory. Try a simpler task.',
+      };
+    }
+
+    // Generic error
+    return {
+      code: 'UNKNOWN_ERROR',
+      message: stderr || `Process exited with code ${exitCode}`,
+      recoverable: false,
+    };
+  }
+
+  /**
+   * Get installation instructions for this CLI
+   * Override to provide CLI-specific instructions
+   */
+  protected getInstallInstructions(): string {
+    const cliName = this.getCliName();
+    const config = this.getSpawnConfig();
+
+    if (process.platform === 'win32') {
+      switch (config.windowsStrategy) {
+        case 'wsl':
+          return `${cliName} requires WSL on Windows. Install WSL, then run inside WSL to install.`;
+        case 'npx':
+          return `Install with: npm install -g ${config.npxPackage || cliName}`;
+        case 'cmd':
+        case 'direct':
+          return `${cliName} is not installed. Check the documentation for installation instructions.`;
+      }
+    }
+
+    return `${cliName} is not installed. Check the documentation for installation instructions.`;
+  }
+
+  // ==========================================================================
+  // CLI Detection
+  // ==========================================================================
+
+  /**
+   * Expand ~ to home directory in path
+   */
+  private expandPath(p: string): string {
+    if (p.startsWith('~')) {
+      return path.join(os.homedir(), p.slice(1));
+    }
+    return p;
+  }
+
+  /**
+   * Find CLI in PATH using 'which' (Unix) or 'where' (Windows)
+   */
+  private findCliInPath(): string | null {
+    const cliName = this.getCliName();
+
+    try {
+      const command = process.platform === 'win32' ? 'where' : 'which';
+      const result = execSync(`${command} ${cliName}`, {
+        encoding: 'utf8',
+        timeout: 5000,
+        stdio: ['pipe', 'pipe', 'pipe'],
+        windowsHide: true,
+      })
+        .trim()
+        .split('\n')[0];
+
+      if (result && fs.existsSync(result)) {
+        cliLogger.debug(`Found ${cliName} in PATH: ${result}`);
+        return result;
+      }
+    } catch {
+      // Not in PATH
+    }
+
+    return null;
+  }
+
+  /**
+   * Find CLI in common installation paths for current platform
+   */
+  private findCliInCommonPaths(): string | null {
+    const config = this.getSpawnConfig();
+    const cliName = this.getCliName();
+    const platform = process.platform as 'linux' | 'darwin' | 'win32';
+    const paths = config.commonPaths[platform] || [];
+
+    for (const p of paths) {
+      const expandedPath = this.expandPath(p);
+      if (fs.existsSync(expandedPath)) {
+        cliLogger.debug(`Found ${cliName} at: ${expandedPath}`);
+        return expandedPath;
+      }
+    }
+
+    return null;
+  }
+
+  /**
+   * Detect CLI installation using appropriate strategy
+   */
+  protected detectCli(): CliDetectionResult {
+    const config = this.getSpawnConfig();
+    const cliName = this.getCliName();
+    const wslLogger = (msg: string) => cliLogger.debug(msg);
+
+    // Windows - use configured strategy
+    if (process.platform === 'win32') {
+      switch (config.windowsStrategy) {
+        case 'wsl': {
+          // Check WSL for CLI
+          if (isWslAvailable({ logger: wslLogger })) {
+            const wslResult: WslCliResult | null = findCliInWsl(cliName, {
+              logger: wslLogger,
+              distribution: config.wslDistribution,
+            });
+            if (wslResult) {
+              cliLogger.debug(
+                `Using ${cliName} via WSL (${wslResult.distribution || 'default'}): ${wslResult.wslPath}`
+              );
+              return {
+                cliPath: 'wsl.exe',
+                useWsl: true,
+                wslCliPath: wslResult.wslPath,
+                wslDistribution: wslResult.distribution,
+                strategy: 'wsl',
+              };
+            }
+          }
+          cliLogger.debug(`${cliName} not found (WSL not available or CLI not installed in WSL)`);
+          return { cliPath: null, useWsl: false, strategy: 'wsl' };
+        }
+
+        case 'npx': {
+          // For npx, we don't need to find the CLI, just return npx
+          cliLogger.debug(`Using ${cliName} via npx (package: ${config.npxPackage})`);
+          return {
+            cliPath: 'npx',
+            useWsl: false,
+            strategy: 'npx',
+          };
+        }
+
+        case 'direct':
+        case 'cmd': {
+          // Native Windows - check PATH and common paths
+          const pathResult = this.findCliInPath();
+          if (pathResult) {
+            return { cliPath: pathResult, useWsl: false, strategy: config.windowsStrategy };
+          }
+
+          const commonResult = this.findCliInCommonPaths();
+          if (commonResult) {
+            return { cliPath: commonResult, useWsl: false, strategy: config.windowsStrategy };
+          }
+
+          cliLogger.debug(`${cliName} not found on Windows`);
+          return { cliPath: null, useWsl: false, strategy: config.windowsStrategy };
+        }
+      }
+    }
+
+    // Linux/macOS - native execution
+    const pathResult = this.findCliInPath();
+    if (pathResult) {
+      return { cliPath: pathResult, useWsl: false, strategy: 'native' };
+    }
+
+    const commonResult = this.findCliInCommonPaths();
+    if (commonResult) {
+      return { cliPath: commonResult, useWsl: false, strategy: 'native' };
+    }
+
+    cliLogger.debug(`${cliName} not found`);
+    return { cliPath: null, useWsl: false, strategy: 'native' };
+  }
+
+  /**
+   * Ensure CLI is detected (lazy initialization)
+   */
+  protected ensureCliDetected(): void {
+    if (this.cliPath !== null || this.detectedStrategy !== 'native') {
+      return; // Already detected
+    }
+
+    const result = this.detectCli();
+    this.cliPath = result.cliPath;
+    this.useWsl = result.useWsl;
+    this.wslCliPath = result.wslCliPath || null;
+    this.wslDistribution = result.wslDistribution;
+    this.detectedStrategy = result.strategy;
+
+    // Set up npx args if using npx strategy
+    const config = this.getSpawnConfig();
+    if (result.strategy === 'npx' && config.npxPackage) {
+      this.npxArgs = [config.npxPackage];
+    }
+  }
+
+  /**
+   * Check if CLI is installed
+   */
+  async isInstalled(): Promise<boolean> {
+    this.ensureCliDetected();
+    return this.cliPath !== null;
+  }
+
+  // ==========================================================================
+  // Subprocess Spawning
+  // ==========================================================================
+
+  /**
+   * Build subprocess options based on detected strategy
+   */
+  protected buildSubprocessOptions(options: ExecuteOptions, cliArgs: string[]): SubprocessOptions {
+    this.ensureCliDetected();
+
+    if (!this.cliPath) {
+      throw new Error(`${this.getCliName()} CLI not found. ${this.getInstallInstructions()}`);
+    }
+
+    const cwd = options.cwd || process.cwd();
+
+    // Filter undefined values from process.env
+    const filteredEnv: Record<string, string> = {};
+    for (const [key, value] of Object.entries(process.env)) {
+      if (value !== undefined) {
+        filteredEnv[key] = value;
+      }
+    }
+
+    // WSL strategy
+    if (this.useWsl && this.wslCliPath) {
+      const wslCwd = windowsToWslPath(cwd);
+      const wslCmd = createWslCommand(this.wslCliPath, cliArgs, {
+        distribution: this.wslDistribution,
+      });
+
+      // Add --cd flag to change directory inside WSL
+      let args: string[];
+      if (this.wslDistribution) {
+        args = ['-d', this.wslDistribution, '--cd', wslCwd, this.wslCliPath, ...cliArgs];
+      } else {
+        args = ['--cd', wslCwd, this.wslCliPath, ...cliArgs];
+      }
+
+      cliLogger.debug(`WSL spawn: ${wslCmd.command} ${args.slice(0, 6).join(' ')}...`);
+
+      return {
+        command: wslCmd.command,
+        args,
+        cwd, // Windows cwd for spawn
+        env: filteredEnv,
+        abortController: options.abortController,
+        timeout: 120000, // CLI operations may take longer
+      };
+    }
+
+    // NPX strategy
+    if (this.detectedStrategy === 'npx') {
+      const allArgs = [...this.npxArgs, ...cliArgs];
+      cliLogger.debug(`NPX spawn: npx ${allArgs.slice(0, 6).join(' ')}...`);
+
+      return {
+        command: 'npx',
+        args: allArgs,
+        cwd,
+        env: filteredEnv,
+        abortController: options.abortController,
+        timeout: 120000,
+      };
+    }
+
+    // Direct strategy (native Unix or Windows direct/cmd)
+    cliLogger.debug(`Direct spawn: ${this.cliPath} ${cliArgs.slice(0, 6).join(' ')}...`);
+
+    return {
+      command: this.cliPath,
+      args: cliArgs,
+      cwd,
+      env: filteredEnv,
+      abortController: options.abortController,
+      timeout: 120000,
+    };
+  }
+
+  /**
+   * Execute a query using the CLI with JSONL streaming
+   *
+   * This is a default implementation that:
+   * 1. Builds CLI args from options
+   * 2. Spawns the subprocess with appropriate strategy
+   * 3. Streams and normalizes events
+   *
+   * Subclasses can override for custom behavior.
+   */
+  async *executeQuery(options: ExecuteOptions): AsyncGenerator<ProviderMessage> {
+    this.ensureCliDetected();
+
+    if (!this.cliPath) {
+      throw new Error(`${this.getCliName()} CLI not found. ${this.getInstallInstructions()}`);
+    }
+
+    const cliArgs = this.buildCliArgs(options);
+    const subprocessOptions = this.buildSubprocessOptions(options, cliArgs);
+
+    try {
+      for await (const rawEvent of spawnJSONLProcess(subprocessOptions)) {
+        const normalized = this.normalizeEvent(rawEvent);
+        if (normalized) {
+          yield normalized;
+        }
+      }
+    } catch (error) {
+      if (isAbortError(error)) {
+        cliLogger.debug('Query aborted');
+        return;
+      }
+
+      // Map CLI errors
+      if (error instanceof Error && 'stderr' in error) {
+        const errorInfo = this.mapError(
+          (error as { stderr?: string }).stderr || error.message,
+          (error as { exitCode?: number | null }).exitCode ?? null
+        );
+
+        const cliError = new Error(errorInfo.message) as Error & CliErrorInfo;
+        cliError.code = errorInfo.code;
+        cliError.recoverable = errorInfo.recoverable;
+        cliError.suggestion = errorInfo.suggestion;
+        throw cliError;
+      }
+
+      throw error;
+    }
+  }
+}

apps/server/src/providers/codex-config-manager.ts

@@ -0,0 +1,85 @@
+/**
+ * Codex Config Manager - Writes MCP server configuration for Codex CLI
+ */
+
+import path from 'path';
+import type { McpServerConfig } from '@automaker/types';
+import * as secureFs from '../lib/secure-fs.js';
+
+const CODEX_CONFIG_DIR = '.codex';
+const CODEX_CONFIG_FILENAME = 'config.toml';
+const CODEX_MCP_SECTION = 'mcp_servers';
+
+function formatTomlString(value: string): string {
+  return JSON.stringify(value);
+}
+
+function formatTomlArray(values: string[]): string {
+  const formatted = values.map((value) => formatTomlString(value)).join(', ');
+  return `[${formatted}]`;
+}
+
+function formatTomlInlineTable(values: Record<string, string>): string {
+  const entries = Object.entries(values).map(
+    ([key, value]) => `${key} = ${formatTomlString(value)}`
+  );
+  return `{ ${entries.join(', ')} }`;
+}
+
+function formatTomlKey(key: string): string {
+  return `"${key.replace(/"/g, '\\"')}"`;
+}
+
+function buildServerBlock(name: string, server: McpServerConfig): string[] {
+  const lines: string[] = [];
+  const section = `${CODEX_MCP_SECTION}.${formatTomlKey(name)}`;
+  lines.push(`[${section}]`);
+
+  if (server.type) {
+    lines.push(`type = ${formatTomlString(server.type)}`);
+  }
+
+  if ('command' in server && server.command) {
+    lines.push(`command = ${formatTomlString(server.command)}`);
+  }
+
+  if ('args' in server && server.args && server.args.length > 0) {
+    lines.push(`args = ${formatTomlArray(server.args)}`);
+  }
+
+  if ('env' in server && server.env && Object.keys(server.env).length > 0) {
+    lines.push(`env = ${formatTomlInlineTable(server.env)}`);
+  }
+
+  if ('url' in server && server.url) {
+    lines.push(`url = ${formatTomlString(server.url)}`);
+  }
+
+  if ('headers' in server && server.headers && Object.keys(server.headers).length > 0) {
+    lines.push(`headers = ${formatTomlInlineTable(server.headers)}`);
+  }
+
+  return lines;
+}
+
+export class CodexConfigManager {
+  async configureMcpServers(
+    cwd: string,
+    mcpServers: Record<string, McpServerConfig>
+  ): Promise<void> {
+    const configDir = path.join(cwd, CODEX_CONFIG_DIR);
+    const configPath = path.join(configDir, CODEX_CONFIG_FILENAME);
+
+    await secureFs.mkdir(configDir, { recursive: true });
+
+    const blocks: string[] = [];
+    for (const [name, server] of Object.entries(mcpServers)) {
+      blocks.push(...buildServerBlock(name, server), '');
+    }
+
+    const content = blocks.join('\n').trim();
+    if (content) {
+      await secureFs.writeFile(configPath, content + '\n', 'utf-8');
+    }
+  }
+}

apps/server/src/providers/codex-models.ts

@@ -0,0 +1,111 @@
+/**
+ * Codex Model Definitions
+ *
+ * Official Codex CLI models as documented at https://developers.openai.com/codex/models/
+ */
+
+import { CODEX_MODEL_MAP } from '@automaker/types';
+import type { ModelDefinition } from './types.js';
+
+const CONTEXT_WINDOW_256K = 256000;
+const CONTEXT_WINDOW_128K = 128000;
+const MAX_OUTPUT_32K = 32000;
+const MAX_OUTPUT_16K = 16000;
+
+/**
+ * All available Codex models with their specifications
+ * Based on https://developers.openai.com/codex/models/
+ */
+export const CODEX_MODELS: ModelDefinition[] = [
+  // ========== Recommended Codex Models ==========
+  {
+    id: CODEX_MODEL_MAP.gpt52Codex,
+    name: 'GPT-5.2-Codex',
+    modelString: CODEX_MODEL_MAP.gpt52Codex,
+    provider: 'openai',
+    description:
+      'Most advanced agentic coding model for complex software engineering (default for ChatGPT users).',
+    contextWindow: CONTEXT_WINDOW_256K,
+    maxOutputTokens: MAX_OUTPUT_32K,
+    supportsVision: true,
+    supportsTools: true,
+    tier: 'premium' as const,
+    default: true,
+    hasReasoning: true,
+  },
+  {
+    id: CODEX_MODEL_MAP.gpt51CodexMax,
+    name: 'GPT-5.1-Codex-Max',
+    modelString: CODEX_MODEL_MAP.gpt51CodexMax,
+    provider: 'openai',
+    description: 'Optimized for long-horizon, agentic coding tasks in Codex.',
+    contextWindow: CONTEXT_WINDOW_256K,
+    maxOutputTokens: MAX_OUTPUT_32K,
+    supportsVision: true,
+    supportsTools: true,
+    tier: 'premium' as const,
+    hasReasoning: true,
+  },
+  {
+    id: CODEX_MODEL_MAP.gpt51CodexMini,
+    name: 'GPT-5.1-Codex-Mini',
+    modelString: CODEX_MODEL_MAP.gpt51CodexMini,
+    provider: 'openai',
+    description: 'Smaller, more cost-effective version for faster workflows.',
+    contextWindow: CONTEXT_WINDOW_128K,
+    maxOutputTokens: MAX_OUTPUT_16K,
+    supportsVision: true,
+    supportsTools: true,
+    tier: 'basic' as const,
+    hasReasoning: false,
+  },
+
+  // ========== General-Purpose GPT Models ==========
+  {
+    id: CODEX_MODEL_MAP.gpt52,
+    name: 'GPT-5.2',
+    modelString: CODEX_MODEL_MAP.gpt52,
+    provider: 'openai',
+    description: 'Best general agentic model for tasks across industries and domains.',
+    contextWindow: CONTEXT_WINDOW_256K,
+    maxOutputTokens: MAX_OUTPUT_32K,
+    supportsVision: true,
+    supportsTools: true,
+    tier: 'standard' as const,
+    hasReasoning: true,
+  },
+  {
+    id: CODEX_MODEL_MAP.gpt51,
+    name: 'GPT-5.1',
+    modelString: CODEX_MODEL_MAP.gpt51,
+    provider: 'openai',
+    description: 'Great for coding and agentic tasks across domains.',
+    contextWindow: CONTEXT_WINDOW_256K,
+    maxOutputTokens: MAX_OUTPUT_32K,
+    supportsVision: true,
+    supportsTools: true,
+    tier: 'standard' as const,
+    hasReasoning: true,
+  },
+];
+
+/**
+ * Get model definition by ID
+ */
+export function getCodexModelById(modelId: string): ModelDefinition | undefined {
+  return CODEX_MODELS.find((m) => m.id === modelId || m.modelString === modelId);
+}
+
+/**
+ * Get all models that support reasoning
+ */
+export function getReasoningModels(): ModelDefinition[] {
+  return CODEX_MODELS.filter((m) => m.hasReasoning);
+}
+
+/**
+ * Get models by tier
+ */
+export function getModelsByTier(tier: 'premium' | 'standard' | 'basic'): ModelDefinition[] {
+  return CODEX_MODELS.filter((m) => m.tier === tier);
+}

apps/server/src/providers/codex-sdk-client.ts

@@ -0,0 +1,173 @@
+/**
+ * Codex SDK client - Executes Codex queries via official @openai/codex-sdk
+ *
+ * Used for programmatic control of Codex from within the application.
+ * Provides cleaner integration than spawning CLI processes.
+ */
+
+import { Codex } from '@openai/codex-sdk';
+import { formatHistoryAsText, classifyError, getUserFriendlyErrorMessage } from '@automaker/utils';
+import { supportsReasoningEffort } from '@automaker/types';
+import type { ExecuteOptions, ProviderMessage } from './types.js';
+
+const OPENAI_API_KEY_ENV = 'OPENAI_API_KEY';
+const SDK_HISTORY_HEADER = 'Current request:\n';
+const DEFAULT_RESPONSE_TEXT = '';
+const SDK_ERROR_DETAILS_LABEL = 'Details:';
+
+type PromptBlock = {
+  type: string;
+  text?: string;
+  source?: {
+    type?: string;
+    media_type?: string;
+    data?: string;
+  };
+};
+
+function resolveApiKey(): string {
+  const apiKey = process.env[OPENAI_API_KEY_ENV];
+  if (!apiKey) {
+    throw new Error('OPENAI_API_KEY is not set.');
+  }
+  return apiKey;
+}
+
+function normalizePromptBlocks(prompt: ExecuteOptions['prompt']): PromptBlock[] {
+  if (Array.isArray(prompt)) {
+    return prompt as PromptBlock[];
+  }
+  return [{ type: 'text', text: prompt }];
+}
+
+function buildPromptText(options: ExecuteOptions, systemPrompt: string | null): string {
+  const historyText =
+    options.conversationHistory && options.conversationHistory.length > 0
+      ? formatHistoryAsText(options.conversationHistory)
+      : '';
+
+  const promptBlocks = normalizePromptBlocks(options.prompt);
+  const promptTexts: string[] = [];
+
+  for (const block of promptBlocks) {
+    if (block.type === 'text' && typeof block.text === 'string' && block.text.trim()) {
+      promptTexts.push(block.text);
+    }
+  }
+
+  const promptContent = promptTexts.join('\n\n');
+  if (!promptContent.trim()) {
+    throw new Error('Codex SDK prompt is empty.');
+  }
+
+  const parts: string[] = [];
+  if (systemPrompt) {
+    parts.push(`System: ${systemPrompt}`);
+  }
+  if (historyText) {
+    parts.push(historyText);
+  }
+  parts.push(`${SDK_HISTORY_HEADER}${promptContent}`);
+
+  return parts.join('\n\n');
+}
+
+function buildSdkErrorMessage(rawMessage: string, userMessage: string): string {
+  if (!rawMessage) {
+    return userMessage;
+  }
+  if (!userMessage || rawMessage === userMessage) {
+    return rawMessage;
+  }
+  return `${userMessage}\n\n${SDK_ERROR_DETAILS_LABEL} ${rawMessage}`;
+}
+
+/**
+ * Execute a query using the official Codex SDK
+ *
+ * The SDK provides a cleaner interface than spawning CLI processes:
+ * - Handles authentication automatically
+ * - Provides TypeScript types
+ * - Supports thread management and resumption
+ * - Better error handling
+ */
+export async function* executeCodexSdkQuery(
+  options: ExecuteOptions,
+  systemPrompt: string | null
+): AsyncGenerator<ProviderMessage> {
+  try {
+    const apiKey = resolveApiKey();
+    const codex = new Codex({ apiKey });
+
+    // Resume existing thread or start new one
+    let thread;
+    if (options.sdkSessionId) {
+      try {
+        thread = codex.resumeThread(options.sdkSessionId);
+      } catch {
+        // If resume fails, start a new thread
+        thread = codex.startThread();
+      }
+    } else {
+      thread = codex.startThread();
+    }
+
+    const promptText = buildPromptText(options, systemPrompt);
+
+    // Build run options with reasoning effort if supported
+    const runOptions: {
+      signal?: AbortSignal;
+      reasoning?: { effort: string };
+    } = {
+      signal: options.abortController?.signal,
+    };
+
+    // Add reasoning effort if model supports it and reasoningEffort is specified
+    if (
+      options.reasoningEffort &&
+      supportsReasoningEffort(options.model) &&
+      options.reasoningEffort !== 'none'
+    ) {
+      runOptions.reasoning = { effort: options.reasoningEffort };
+    }
+
+    // Run the query
+    const result = await thread.run(promptText, runOptions);
+
+    // Extract response text (from finalResponse property)
+    const outputText = result.finalResponse ?? DEFAULT_RESPONSE_TEXT;
+
+    // Get thread ID (may be null if not populated yet)
+    const threadId = thread.id ?? undefined;
+
+    // Yield assistant message
+    yield {
+      type: 'assistant',
+      session_id: threadId,
+      message: {
+        role: 'assistant',
+        content: [{ type: 'text', text: outputText }],
+      },
+    };
+
+    // Yield result
+    yield {
+      type: 'result',
+      subtype: 'success',
+      session_id: threadId,
+      result: outputText,
+    };
+  } catch (error) {
+    const errorInfo = classifyError(error);
+    const userMessage = getUserFriendlyErrorMessage(error);
+    const combinedMessage = buildSdkErrorMessage(errorInfo.message, userMessage);
+    console.error('[CodexSDK] executeQuery() error during execution:', {
+      type: errorInfo.type,
+      message: errorInfo.message,
+      isRateLimit: errorInfo.isRateLimit,
+      retryAfter: errorInfo.retryAfter,
+      stack: error instanceof Error ? error.stack : undefined,
+    });
+    yield { type: 'error', error: combinedMessage };
+  }
+}

apps/server/src/providers/codex-tool-mapping.ts

@@ -0,0 +1,436 @@
+export type CodexToolResolution = {
+  name: string;
+  input: Record<string, unknown>;
+};
+
+export type CodexTodoItem = {
+  content: string;
+  status: 'pending' | 'in_progress' | 'completed';
+  activeForm?: string;
+};
+
+const TOOL_NAME_BASH = 'Bash';
+const TOOL_NAME_READ = 'Read';
+const TOOL_NAME_EDIT = 'Edit';
+const TOOL_NAME_WRITE = 'Write';
+const TOOL_NAME_GREP = 'Grep';
+const TOOL_NAME_GLOB = 'Glob';
+const TOOL_NAME_TODO = 'TodoWrite';
+const TOOL_NAME_DELETE = 'Delete';
+const TOOL_NAME_LS = 'Ls';
+
+const INPUT_KEY_COMMAND = 'command';
+const INPUT_KEY_FILE_PATH = 'file_path';
+const INPUT_KEY_PATTERN = 'pattern';
+
+const SHELL_WRAPPER_PATTERNS = [
+  /^\/bin\/bash\s+-lc\s+["']([\s\S]+)["']$/,
+  /^bash\s+-lc\s+["']([\s\S]+)["']$/,
+  /^\/bin\/sh\s+-lc\s+["']([\s\S]+)["']$/,
+  /^sh\s+-lc\s+["']([\s\S]+)["']$/,
+  /^cmd\.exe\s+\/c\s+["']?([\s\S]+)["']?$/i,
+  /^powershell(?:\.exe)?\s+-Command\s+["']?([\s\S]+)["']?$/i,
+  /^pwsh(?:\.exe)?\s+-Command\s+["']?([\s\S]+)["']?$/i,
+] as const;
+
+const COMMAND_SEPARATOR_PATTERN = /\s*(?:&&|\|\||;)\s*/;
+const SEGMENT_SKIP_PREFIXES = ['cd ', 'export ', 'set ', 'pushd '] as const;
+const WRAPPER_COMMANDS = new Set(['sudo', 'env', 'command']);
+const READ_COMMANDS = new Set(['cat', 'sed', 'head', 'tail', 'less', 'more', 'bat', 'stat', 'wc']);
+const SEARCH_COMMANDS = new Set(['rg', 'grep', 'ag', 'ack']);
+const GLOB_COMMANDS = new Set(['ls', 'find', 'fd', 'tree']);
+const DELETE_COMMANDS = new Set(['rm', 'del', 'erase', 'remove', 'unlink']);
+const LIST_COMMANDS = new Set(['ls', 'dir', 'll', 'la']);
+const WRITE_COMMANDS = new Set(['tee', 'touch', 'mkdir']);
+const APPLY_PATCH_COMMAND = 'apply_patch';
+const APPLY_PATCH_PATTERN = /\bapply_patch\b/;
+const REDIRECTION_TARGET_PATTERN = /(?:>>|>)\s*([^\s]+)/;
+const SED_IN_PLACE_FLAGS = new Set(['-i', '--in-place']);
+const PERL_IN_PLACE_FLAG = /-.*i/;
+const SEARCH_PATTERN_FLAGS = new Set(['-e', '--regexp']);
+const SEARCH_VALUE_FLAGS = new Set([
+  '-g',
+  '--glob',
+  '--iglob',
+  '--type',
+  '--type-add',
+  '--type-clear',
+  '--encoding',
+]);
+const SEARCH_FILE_LIST_FLAGS = new Set(['--files']);
+const TODO_LINE_PATTERN = /^[-*]\s*(?:\[(?<status>[ x~])\]\s*)?(?<content>.+)$/;
+const TODO_STATUS_COMPLETED = 'completed';
+const TODO_STATUS_IN_PROGRESS = 'in_progress';
+const TODO_STATUS_PENDING = 'pending';
+const PATCH_FILE_MARKERS = [
+  '*** Update File: ',
+  '*** Add File: ',
+  '*** Delete File: ',
+  '*** Move to: ',
+] as const;
+
+function stripShellWrapper(command: string): string {
+  const trimmed = command.trim();
+  for (const pattern of SHELL_WRAPPER_PATTERNS) {
+    const match = trimmed.match(pattern);
+    if (match && match[1]) {
+      return unescapeCommand(match[1].trim());
+    }
+  }
+  return trimmed;
+}
+
+function unescapeCommand(command: string): string {
+  return command.replace(/\\(["'])/g, '$1');
+}
+
+function extractPrimarySegment(command: string): string {
+  const segments = command
+    .split(COMMAND_SEPARATOR_PATTERN)
+    .map((segment) => segment.trim())
+    .filter(Boolean);
+
+  for (const segment of segments) {
+    const shouldSkip = SEGMENT_SKIP_PREFIXES.some((prefix) => segment.startsWith(prefix));
+    if (!shouldSkip) {
+      return segment;
+    }
+  }
+
+  return command.trim();
+}
+
+function tokenizeCommand(command: string): string[] {
+  const tokens: string[] = [];
+  let current = '';
+  let inSingleQuote = false;
+  let inDoubleQuote = false;
+  let isEscaped = false;
+
+  for (const char of command) {
+    if (isEscaped) {
+      current += char;
+      isEscaped = false;
+      continue;
+    }
+
+    if (char === '\\') {
+      isEscaped = true;
+      continue;
+    }
+
+    if (char === "'" && !inDoubleQuote) {
+      inSingleQuote = !inSingleQuote;
+      continue;
+    }
+
+    if (char === '"' && !inSingleQuote) {
+      inDoubleQuote = !inDoubleQuote;
+      continue;
+    }
+
+    if (!inSingleQuote && !inDoubleQuote && /\s/.test(char)) {
+      if (current) {
+        tokens.push(current);
+        current = '';
+      }
+      continue;
+    }
+
+    current += char;
+  }
+
+  if (current) {
+    tokens.push(current);
+  }
+
+  return tokens;
+}
+
+function stripWrapperTokens(tokens: string[]): string[] {
+  let index = 0;
+  while (index < tokens.length && WRAPPER_COMMANDS.has(tokens[index].toLowerCase())) {
+    index += 1;
+  }
+  return tokens.slice(index);
+}
+
+function extractFilePathFromTokens(tokens: string[]): string | null {
+  const candidates = tokens.slice(1).filter((token) => token && !token.startsWith('-'));
+  if (candidates.length === 0) return null;
+  return candidates[candidates.length - 1];
+}
+
+function extractSearchPattern(tokens: string[]): string | null {
+  const remaining = tokens.slice(1);
+
+  for (let index = 0; index < remaining.length; index += 1) {
+    const token = remaining[index];
+    if (token === '--') {
+      return remaining[index + 1] ?? null;
+    }
+    if (SEARCH_PATTERN_FLAGS.has(token)) {
+      return remaining[index + 1] ?? null;
+    }
+    if (SEARCH_VALUE_FLAGS.has(token)) {
+      index += 1;
+      continue;
+    }
+    if (token.startsWith('-')) {
+      continue;
+    }
+    return token;
+  }
+
+  return null;
+}
+
+function extractTeeTarget(tokens: string[]): string | null {
+  const teeIndex = tokens.findIndex((token) => token === 'tee');
+  if (teeIndex < 0) return null;
+  const candidate = tokens[teeIndex + 1];
+  return candidate && !candidate.startsWith('-') ? candidate : null;
+}
+
+function extractRedirectionTarget(command: string): string | null {
+  const match = command.match(REDIRECTION_TARGET_PATTERN);
+  return match?.[1] ?? null;
+}
+
+function extractFilePathFromDeleteTokens(tokens: string[]): string | null {
+  // rm file.txt or rm /path/to/file.txt
+  // Skip flags and get the first non-flag argument
+  for (let i = 1; i < tokens.length; i++) {
+    const token = tokens[i];
+    if (token && !token.startsWith('-')) {
+      return token;
+    }
+  }
+  return null;
+}
+
+function hasSedInPlaceFlag(tokens: string[]): boolean {
+  return tokens.some((token) => SED_IN_PLACE_FLAGS.has(token) || token.startsWith('-i'));
+}
+
+function hasPerlInPlaceFlag(tokens: string[]): boolean {
+  return tokens.some((token) => PERL_IN_PLACE_FLAG.test(token));
+}
+
+function extractPatchFilePath(command: string): string | null {
+  for (const marker of PATCH_FILE_MARKERS) {
+    const index = command.indexOf(marker);
+    if (index < 0) continue;
+    const start = index + marker.length;
+    const end = command.indexOf('\n', start);
+    const rawPath = (end === -1 ? command.slice(start) : command.slice(start, end)).trim();
+    if (rawPath) return rawPath;
+  }
+  return null;
+}
+
+function buildInputWithFilePath(filePath: string | null): Record<string, unknown> {
+  return filePath ? { [INPUT_KEY_FILE_PATH]: filePath } : {};
+}
+
+function buildInputWithPattern(pattern: string | null): Record<string, unknown> {
+  return pattern ? { [INPUT_KEY_PATTERN]: pattern } : {};
+}
+
+export function resolveCodexToolCall(command: string): CodexToolResolution {
+  const normalized = stripShellWrapper(command);
+  const primarySegment = extractPrimarySegment(normalized);
+  const tokens = stripWrapperTokens(tokenizeCommand(primarySegment));
+  const commandToken = tokens[0]?.toLowerCase() ?? '';
+
+  const redirectionTarget = extractRedirectionTarget(primarySegment);
+  if (redirectionTarget) {
+    return {
+      name: TOOL_NAME_WRITE,
+      input: buildInputWithFilePath(redirectionTarget),
+    };
+  }
+
+  if (commandToken === APPLY_PATCH_COMMAND || APPLY_PATCH_PATTERN.test(primarySegment)) {
+    return {
+      name: TOOL_NAME_EDIT,
+      input: buildInputWithFilePath(extractPatchFilePath(primarySegment)),
+    };
+  }
+
+  if (commandToken === 'sed' && hasSedInPlaceFlag(tokens)) {
+    return {
+      name: TOOL_NAME_EDIT,
+      input: buildInputWithFilePath(extractFilePathFromTokens(tokens)),
+    };
+  }
+
+  if (commandToken === 'perl' && hasPerlInPlaceFlag(tokens)) {
+    return {
+      name: TOOL_NAME_EDIT,
+      input: buildInputWithFilePath(extractFilePathFromTokens(tokens)),
+    };
+  }
+
+  if (WRITE_COMMANDS.has(commandToken)) {
+    const filePath =
+      commandToken === 'tee' ? extractTeeTarget(tokens) : extractFilePathFromTokens(tokens);
+    return {
+      name: TOOL_NAME_WRITE,
+      input: buildInputWithFilePath(filePath),
+    };
+  }
+
+  if (SEARCH_COMMANDS.has(commandToken)) {
+    if (tokens.some((token) => SEARCH_FILE_LIST_FLAGS.has(token))) {
+      return {
+        name: TOOL_NAME_GLOB,
+        input: buildInputWithPattern(extractFilePathFromTokens(tokens)),
+      };
+    }
+
+    return {
+      name: TOOL_NAME_GREP,
+      input: buildInputWithPattern(extractSearchPattern(tokens)),
+    };
+  }
+
+  // Handle Delete commands (rm, del, erase, remove, unlink)
+  if (DELETE_COMMANDS.has(commandToken)) {
+    // Skip if -r or -rf flags (recursive delete should go to Bash)
+    if (
+      tokens.some((token) => token === '-r' || token === '-rf' || token === '-f' || token === '-rf')
+    ) {
+      return {
+        name: TOOL_NAME_BASH,
+        input: { [INPUT_KEY_COMMAND]: normalized },
+      };
+    }
+    // Simple file deletion - extract the file path
+    const filePath = extractFilePathFromDeleteTokens(tokens);
+    if (filePath) {
+      return {
+        name: TOOL_NAME_DELETE,
+        input: { path: filePath },
+      };
+    }
+    // Fall back to bash if we can't determine the file path
+    return {
+      name: TOOL_NAME_BASH,
+      input: { [INPUT_KEY_COMMAND]: normalized },
+    };
+  }
+
+  // Handle simple Ls commands (just listing, not find/glob)
+  if (LIST_COMMANDS.has(commandToken)) {
+    const filePath = extractFilePathFromTokens(tokens);
+    return {
+      name: TOOL_NAME_LS,
+      input: { path: filePath || '.' },
+    };
+  }
+
+  if (GLOB_COMMANDS.has(commandToken)) {
+    return {
+      name: TOOL_NAME_GLOB,
+      input: buildInputWithPattern(extractFilePathFromTokens(tokens)),
+    };
+  }
+
+  if (READ_COMMANDS.has(commandToken)) {
+    return {
+      name: TOOL_NAME_READ,
+      input: buildInputWithFilePath(extractFilePathFromTokens(tokens)),
+    };
+  }
+
+  return {
+    name: TOOL_NAME_BASH,
+    input: { [INPUT_KEY_COMMAND]: normalized },
+  };
+}
+
+function parseTodoLines(lines: string[]): CodexTodoItem[] {
+  const todos: CodexTodoItem[] = [];
+
+  for (const line of lines) {
+    const match = line.match(TODO_LINE_PATTERN);
+    if (!match?.groups?.content) continue;
+
+    const statusToken = match.groups.status;
+    const status =
+      statusToken === 'x'
+        ? TODO_STATUS_COMPLETED
+        : statusToken === '~'
+          ? TODO_STATUS_IN_PROGRESS
+          : TODO_STATUS_PENDING;
+
+    todos.push({ content: match.groups.content.trim(), status });
+  }
+
+  return todos;
+}
+
+function extractTodoFromArray(value: unknown[]): CodexTodoItem[] {
+  return value
+    .map((entry) => {
+      if (typeof entry === 'string') {
+        return { content: entry, status: TODO_STATUS_PENDING };
+      }
+      if (entry && typeof entry === 'object') {
+        const record = entry as Record<string, unknown>;
+        const content =
+          typeof record.content === 'string'
+            ? record.content
+            : typeof record.text === 'string'
+              ? record.text
+              : typeof record.title === 'string'
+                ? record.title
+                : null;
+        if (!content) return null;
+        const status =
+          record.status === TODO_STATUS_COMPLETED ||
+          record.status === TODO_STATUS_IN_PROGRESS ||
+          record.status === TODO_STATUS_PENDING
+            ? (record.status as CodexTodoItem['status'])
+            : TODO_STATUS_PENDING;
+        const activeForm = typeof record.activeForm === 'string' ? record.activeForm : undefined;
+        return { content, status, activeForm };
+      }
+      return null;
+    })
+    .filter((item): item is CodexTodoItem => Boolean(item));
+}
+
+export function extractCodexTodoItems(item: Record<string, unknown>): CodexTodoItem[] | null {
+  const todosValue = item.todos;
+  if (Array.isArray(todosValue)) {
+    const todos = extractTodoFromArray(todosValue);
+    return todos.length > 0 ? todos : null;
+  }
+
+  const itemsValue = item.items;
+  if (Array.isArray(itemsValue)) {
+    const todos = extractTodoFromArray(itemsValue);
+    return todos.length > 0 ? todos : null;
+  }
+
+  const textValue =
+    typeof item.text === 'string'
+      ? item.text
+      : typeof item.content === 'string'
+        ? item.content
+        : null;
+  if (!textValue) return null;
+
+  const lines = textValue
+    .split('\n')
+    .map((line) => line.trim())
+    .filter(Boolean);
+  const todos = parseTodoLines(lines);
+  return todos.length > 0 ? todos : null;
+}
+
+export function getCodexTodoToolName(): string {
+  return TOOL_NAME_TODO;
+}

apps/server/src/providers/cursor-config-manager.ts

@@ -0,0 +1,197 @@
+/**
+ * Cursor CLI Configuration Manager
+ *
+ * Manages Cursor CLI configuration stored in .automaker/cursor-config.json
+ */
+
+import * as fs from 'fs';
+import * as path from 'path';
+import { getAllCursorModelIds, type CursorCliConfig, type CursorModelId } from '@automaker/types';
+import { createLogger } from '@automaker/utils';
+import { getAutomakerDir } from '@automaker/platform';
+
+// Create logger for this module
+const logger = createLogger('CursorConfigManager');
+
+/**
+ * Manages Cursor CLI configuration
+ * Config location: .automaker/cursor-config.json
+ */
+export class CursorConfigManager {
+  private configPath: string;
+  private config: CursorCliConfig;
+
+  constructor(projectPath: string) {
+    // Use getAutomakerDir for consistent path resolution
+    this.configPath = path.join(getAutomakerDir(projectPath), 'cursor-config.json');
+    this.config = this.loadConfig();
+  }
+
+  /**
+   * Load configuration from disk
+   */
+  private loadConfig(): CursorCliConfig {
+    try {
+      if (fs.existsSync(this.configPath)) {
+        const content = fs.readFileSync(this.configPath, 'utf8');
+        const parsed = JSON.parse(content) as CursorCliConfig;
+        logger.debug(`Loaded config from ${this.configPath}`);
+        return parsed;
+      }
+    } catch (error) {
+      logger.warn('Failed to load config:', error);
+    }
+
+    // Return default config with all available models
+    return {
+      defaultModel: 'auto',
+      models: getAllCursorModelIds(),
+    };
+  }
+
+  /**
+   * Save configuration to disk
+   */
+  private saveConfig(): void {
+    try {
+      const dir = path.dirname(this.configPath);
+      if (!fs.existsSync(dir)) {
+        fs.mkdirSync(dir, { recursive: true });
+      }
+      fs.writeFileSync(this.configPath, JSON.stringify(this.config, null, 2));
+      logger.debug('Config saved');
+    } catch (error) {
+      logger.error('Failed to save config:', error);
+      throw error;
+    }
+  }
+
+  /**
+   * Get the full configuration
+   */
+  getConfig(): CursorCliConfig {
+    return { ...this.config };
+  }
+
+  /**
+   * Get the default model
+   */
+  getDefaultModel(): CursorModelId {
+    return this.config.defaultModel || 'auto';
+  }
+
+  /**
+   * Set the default model
+   */
+  setDefaultModel(model: CursorModelId): void {
+    this.config.defaultModel = model;
+    this.saveConfig();
+    logger.info(`Default model set to: ${model}`);
+  }
+
+  /**
+   * Get enabled models
+   */
+  getEnabledModels(): CursorModelId[] {
+    return this.config.models || ['auto'];
+  }
+
+  /**
+   * Set enabled models
+   */
+  setEnabledModels(models: CursorModelId[]): void {
+    this.config.models = models;
+    this.saveConfig();
+    logger.info(`Enabled models updated: ${models.join(', ')}`);
+  }
+
+  /**
+   * Add a model to enabled list
+   */
+  addModel(model: CursorModelId): void {
+    if (!this.config.models) {
+      this.config.models = [];
+    }
+    if (!this.config.models.includes(model)) {
+      this.config.models.push(model);
+      this.saveConfig();
+      logger.info(`Model added: ${model}`);
+    }
+  }
+
+  /**
+   * Remove a model from enabled list
+   */
+  removeModel(model: CursorModelId): void {
+    if (this.config.models) {
+      this.config.models = this.config.models.filter((m) => m !== model);
+      this.saveConfig();
+      logger.info(`Model removed: ${model}`);
+    }
+  }
+
+  /**
+   * Check if a model is enabled
+   */
+  isModelEnabled(model: CursorModelId): boolean {
+    return this.config.models?.includes(model) ?? false;
+  }
+
+  /**
+   * Get MCP server configurations
+   */
+  getMcpServers(): string[] {
+    return this.config.mcpServers || [];
+  }
+
+  /**
+   * Set MCP server configurations
+   */
+  setMcpServers(servers: string[]): void {
+    this.config.mcpServers = servers;
+    this.saveConfig();
+    logger.info(`MCP servers updated: ${servers.join(', ')}`);
+  }
+
+  /**
+   * Get Cursor rules paths
+   */
+  getRules(): string[] {
+    return this.config.rules || [];
+  }
+
+  /**
+   * Set Cursor rules paths
+   */
+  setRules(rules: string[]): void {
+    this.config.rules = rules;
+    this.saveConfig();
+    logger.info(`Rules updated: ${rules.join(', ')}`);
+  }
+
+  /**
+   * Reset configuration to defaults
+   */
+  reset(): void {
+    this.config = {
+      defaultModel: 'auto',
+      models: getAllCursorModelIds(),
+    };
+    this.saveConfig();
+    logger.info('Config reset to defaults');
+  }
+
+  /**
+   * Check if config file exists
+   */
+  exists(): boolean {
+    return fs.existsSync(this.configPath);
+  }
+
+  /**
+   * Get the config file path
+   */
+  getConfigPath(): string {
+    return this.configPath;
+  }
+}

apps/server/src/providers/index.ts

@@ -0,0 +1,40 @@
+/**
+ * Provider exports
+ */
+
+// Base providers
+export { BaseProvider } from './base-provider.js';
+export {
+  CliProvider,
+  type SpawnStrategy,
+  type CliSpawnConfig,
+  type CliErrorInfo,
+} from './cli-provider.js';
+export type {
+  ProviderConfig,
+  ExecuteOptions,
+  ProviderMessage,
+  InstallationStatus,
+  ModelDefinition,
+} from './types.js';
+
+// Claude provider
+export { ClaudeProvider } from './claude-provider.js';
+
+// Cursor provider
+export { CursorProvider, CursorErrorCode, CursorError } from './cursor-provider.js';
+export { CursorConfigManager } from './cursor-config-manager.js';
+
+// OpenCode provider
+export { OpencodeProvider } from './opencode-provider.js';
+
+// Provider factory
+export { ProviderFactory } from './provider-factory.js';
+
+// Simple query service - unified interface for basic AI queries
+export { simpleQuery, streamingQuery } from './simple-query-service.js';
+export type {
+  SimpleQueryOptions,
+  SimpleQueryResult,
+  StreamingQueryOptions,
+} from './simple-query-service.js';

apps/server/src/providers/provider-factory.ts

@@ -1,51 +1,168 @@
 /**
  * Provider Factory - Routes model IDs to the appropriate provider
  *
- * This factory implements model-based routing to automatically select
- * the correct provider based on the model string. This makes adding
- * new providers (Cursor, OpenCode, etc.) trivial - just add one line.
+ * Uses a registry pattern for dynamic provider registration.
+ * Providers register themselves on import, making it easy to add new providers.
  */
 
 import { BaseProvider } from './base-provider.js';
-import { ClaudeProvider } from './claude-provider.js';
-import type { InstallationStatus } from './types.js';
+import type { InstallationStatus, ModelDefinition } from './types.js';
+import { isCursorModel, isCodexModel, isOpencodeModel, type ModelProvider } from '@automaker/types';
+import * as fs from 'fs';
+import * as path from 'path';
+
+const DISCONNECTED_MARKERS: Record<string, string> = {
+  claude: '.claude-disconnected',
+  codex: '.codex-disconnected',
+  cursor: '.cursor-disconnected',
+  opencode: '.opencode-disconnected',
+};
+
+/**
+ * Check if a provider CLI is disconnected from the app
+ */
+export function isProviderDisconnected(providerName: string): boolean {
+  const markerFile = DISCONNECTED_MARKERS[providerName.toLowerCase()];
+  if (!markerFile) return false;
+
+  const markerPath = path.join(process.cwd(), '.automaker', markerFile);
+  return fs.existsSync(markerPath);
+}
+
+/**
+ * Provider registration entry
+ */
+interface ProviderRegistration {
+  /** Factory function to create provider instance */
+  factory: () => BaseProvider;
+  /** Aliases for this provider (e.g., 'anthropic' for 'claude') */
+  aliases?: string[];
+  /** Function to check if this provider can handle a model ID */
+  canHandleModel?: (modelId: string) => boolean;
+  /** Priority for model matching (higher = checked first) */
+  priority?: number;
+}
+
+/**
+ * Provider registry - stores registered providers
+ */
+const providerRegistry = new Map<string, ProviderRegistration>();
+
+/**
+ * Register a provider with the factory
+ *
+ * @param name Provider name (e.g., 'claude', 'cursor')
+ * @param registration Provider registration config
+ */
+export function registerProvider(name: string, registration: ProviderRegistration): void {
+  providerRegistry.set(name.toLowerCase(), registration);
+}
 
 export class ProviderFactory {
   /**
-   * Get the appropriate provider for a given model ID
+   * Determine which provider to use for a given model
    *
-   * @param modelId Model identifier (e.g., "claude-opus-4-5-20251101", "gpt-5.2", "cursor-fast")
-   * @returns Provider instance for the model
+   * @param model Model identifier
+   * @returns Provider name (ModelProvider type)
    */
-  static getProviderForModel(modelId: string): BaseProvider {
-    const lowerModel = modelId.toLowerCase();
+  static getProviderNameForModel(model: string): ModelProvider {
+    const lowerModel = model.toLowerCase();
 
-    // Claude models (claude-*, opus, sonnet, haiku)
-    if (lowerModel.startsWith('claude-') || ['haiku', 'sonnet', 'opus'].includes(lowerModel)) {
-      return new ClaudeProvider();
+    // Get all registered providers sorted by priority (descending)
+    const registrations = Array.from(providerRegistry.entries()).sort(
+      ([, a], [, b]) => (b.priority ?? 0) - (a.priority ?? 0)
+    );
+
+    // Check each provider's canHandleModel function
+    for (const [name, reg] of registrations) {
+      if (reg.canHandleModel?.(lowerModel)) {
+        return name as ModelProvider;
+      }
     }
 
-    // Future providers:
-    // if (lowerModel.startsWith("cursor-")) {
-    //   return new CursorProvider();
-    // }
-    // if (lowerModel.startsWith("opencode-")) {
-    //   return new OpenCodeProvider();
-    // }
+    // Fallback: Check for explicit prefixes
+    for (const [name] of registrations) {
+      if (lowerModel.startsWith(`${name}-`)) {
+        return name as ModelProvider;
+      }
+    }
 
-    // Default to Claude for unknown models
-    console.warn(`[ProviderFactory] Unknown model prefix for "${modelId}", defaulting to Claude`);
-    return new ClaudeProvider();
+    // Default to claude (first registered provider or claude)
+    return 'claude';
+  }
+
+  /**
+   * Get the appropriate provider for a given model ID
+   *
+   * @param modelId Model identifier (e.g., "claude-opus-4-5-20251101", "cursor-gpt-4o", "cursor-auto")
+   * @param options Optional settings
+   * @param options.throwOnDisconnected Throw error if provider is disconnected (default: true)
+   * @returns Provider instance for the model
+   * @throws Error if provider is disconnected and throwOnDisconnected is true
+   */
+  static getProviderForModel(
+    modelId: string,
+    options: { throwOnDisconnected?: boolean } = {}
+  ): BaseProvider {
+    const { throwOnDisconnected = true } = options;
+    const providerName = this.getProviderForModelName(modelId);
+
+    // Check if provider is disconnected
+    if (throwOnDisconnected && isProviderDisconnected(providerName)) {
+      throw new Error(
+        `${providerName.charAt(0).toUpperCase() + providerName.slice(1)} CLI is disconnected from the app. ` +
+          `Please go to Settings > Providers and click "Sign In" to reconnect.`
+      );
+    }
+
+    const provider = this.getProviderByName(providerName);
+
+    if (!provider) {
+      // Fallback to claude if provider not found
+      const claudeReg = providerRegistry.get('claude');
+      if (claudeReg) {
+        return claudeReg.factory();
+      }
+      throw new Error(`No provider found for model: ${modelId}`);
+    }
+
+    return provider;
+  }
+
+  /**
+   * Get the provider name for a given model ID (without creating provider instance)
+   */
+  static getProviderForModelName(modelId: string): string {
+    const lowerModel = modelId.toLowerCase();
+
+    // Get all registered providers sorted by priority (descending)
+    const registrations = Array.from(providerRegistry.entries()).sort(
+      ([, a], [, b]) => (b.priority ?? 0) - (a.priority ?? 0)
+    );
+
+    // Check each provider's canHandleModel function
+    for (const [name, reg] of registrations) {
+      if (reg.canHandleModel?.(lowerModel)) {
+        return name;
+      }
+    }
+
+    // Fallback: Check for explicit prefixes
+    for (const [name] of registrations) {
+      if (lowerModel.startsWith(`${name}-`)) {
+        return name;
+      }
+    }
+
+    // Default to claude (first registered provider or claude)
+    return 'claude';
   }
 
   /**
    * Get all available providers
    */
   static getAllProviders(): BaseProvider[] {
-    return [
-      new ClaudeProvider(),
-      // Future providers...
-    ];
+    return Array.from(providerRegistry.values()).map((reg) => reg.factory());
   }
 
   /**
@@ -54,11 +171,10 @@ export class ProviderFactory {
    * @returns Map of provider name to installation status
    */
   static async checkAllProviders(): Promise<Record<string, InstallationStatus>> {
-    const providers = this.getAllProviders();
     const statuses: Record<string, InstallationStatus> = {};
 
-    for (const provider of providers) {
-      const name = provider.getName();
+    for (const [name, reg] of providerRegistry.entries()) {
+      const provider = reg.factory();
       const status = await provider.detectInstallation();
       statuses[name] = status;
     }
@@ -69,40 +185,119 @@ export class ProviderFactory {
   /**
    * Get provider by name (for direct access if needed)
    *
-   * @param name Provider name (e.g., "claude", "cursor")
+   * @param name Provider name (e.g., "claude", "cursor") or alias (e.g., "anthropic")
    * @returns Provider instance or null if not found
    */
   static getProviderByName(name: string): BaseProvider | null {
     const lowerName = name.toLowerCase();
 
-    switch (lowerName) {
-      case 'claude':
-      case 'anthropic':
-        return new ClaudeProvider();
-
-      // Future providers:
-      // case "cursor":
-      //   return new CursorProvider();
-      // case "opencode":
-      //   return new OpenCodeProvider();
-
-      default:
-        return null;
+    // Direct lookup
+    const directReg = providerRegistry.get(lowerName);
+    if (directReg) {
+      return directReg.factory();
     }
+
+    // Check aliases
+    for (const [, reg] of providerRegistry.entries()) {
+      if (reg.aliases?.includes(lowerName)) {
+        return reg.factory();
+      }
+    }
+
+    return null;
   }
 
   /**
    * Get all available models from all providers
    */
-  static getAllAvailableModels() {
+  static getAllAvailableModels(): ModelDefinition[] {
     const providers = this.getAllProviders();
-    const allModels = [];
+    return providers.flatMap((p) => p.getAvailableModels());
+  }
 
-    for (const provider of providers) {
-      const models = provider.getAvailableModels();
-      allModels.push(...models);
+  /**
+   * Get list of registered provider names
+   */
+  static getRegisteredProviderNames(): string[] {
+    return Array.from(providerRegistry.keys());
+  }
+
+  /**
+   * Check if a specific model supports vision/image input
+   *
+   * @param modelId Model identifier
+   * @returns Whether the model supports vision (defaults to true if model not found)
+   */
+  static modelSupportsVision(modelId: string): boolean {
+    const provider = this.getProviderForModel(modelId);
+    const models = provider.getAvailableModels();
+
+    // Find the model in the available models list
+    for (const model of models) {
+      if (
+        model.id === modelId ||
+        model.modelString === modelId ||
+        model.id.endsWith(`-${modelId}`) ||
+        model.modelString.endsWith(`-${modelId}`) ||
+        model.modelString === modelId.replace(/^(claude|cursor|codex)-/, '') ||
+        model.modelString === modelId.replace(/-(claude|cursor|codex)$/, '')
+      ) {
+        return model.supportsVision ?? true;
+      }
     }
 
-    return allModels;
+    // Also try exact match with model string from provider's model map
+    for (const model of models) {
+      if (model.modelString === modelId || model.id === modelId) {
+        return model.supportsVision ?? true;
+      }
+    }
+
+    // Default to true (Claude SDK supports vision by default)
+    return true;
   }
 }
+
+// =============================================================================
+// Provider Registrations
+// =============================================================================
+
+// Import providers for registration side-effects
+import { ClaudeProvider } from './claude-provider.js';
+import { CursorProvider } from './cursor-provider.js';
+import { CodexProvider } from './codex-provider.js';
+import { OpencodeProvider } from './opencode-provider.js';
+
+// Register Claude provider
+registerProvider('claude', {
+  factory: () => new ClaudeProvider(),
+  aliases: ['anthropic'],
+  canHandleModel: (model: string) => {
+    return (
+      model.startsWith('claude-') || ['opus', 'sonnet', 'haiku'].some((n) => model.includes(n))
+    );
+  },
+  priority: 0, // Default priority
+});
+
+// Register Cursor provider
+registerProvider('cursor', {
+  factory: () => new CursorProvider(),
+  canHandleModel: (model: string) => isCursorModel(model),
+  priority: 10, // Higher priority - check Cursor models first
+});
+
+// Register Codex provider
+registerProvider('codex', {
+  factory: () => new CodexProvider(),
+  aliases: ['openai'],
+  canHandleModel: (model: string) => isCodexModel(model),
+  priority: 5, // Medium priority - check after Cursor but before Claude
+});
+
+// Register OpenCode provider
+registerProvider('opencode', {
+  factory: () => new OpencodeProvider(),
+  canHandleModel: (model: string) => isOpencodeModel(model),
+  priority: 3, // Between codex (5) and claude (0)
+});

apps/server/src/providers/simple-query-service.ts

@@ -0,0 +1,240 @@
+/**
+ * Simple Query Service - Simplified interface for basic AI queries
+ *
+ * Use this for routes that need simple text responses without
+ * complex event handling. This service abstracts away the provider
+ * selection and streaming details, providing a clean interface
+ * for common query patterns.
+ *
+ * Benefits:
+ * - No direct SDK imports needed in route files
+ * - Consistent provider routing based on model
+ * - Automatic text extraction from streaming responses
+ * - Structured output support for JSON schema responses
+ * - Eliminates duplicate extractTextFromStream() functions
+ */
+
+import { ProviderFactory } from './provider-factory.js';
+import type { ProviderMessage, ContentBlock, ThinkingLevel } from '@automaker/types';
+
+/**
+ * Options for simple query execution
+ */
+export interface SimpleQueryOptions {
+  /** The prompt to send to the AI (can be text or multi-part content) */
+  prompt: string | Array<{ type: string; text?: string; source?: object }>;
+  /** Model to use (with or without provider prefix) */
+  model?: string;
+  /** Working directory for the query */
+  cwd: string;
+  /** System prompt (combined with user prompt for some providers) */
+  systemPrompt?: string;
+  /** Maximum turns for agentic operations (default: 1) */
+  maxTurns?: number;
+  /** Tools to allow (default: [] for simple queries) */
+  allowedTools?: string[];
+  /** Abort controller for cancellation */
+  abortController?: AbortController;
+  /** Structured output format for JSON responses */
+  outputFormat?: {
+    type: 'json_schema';
+    schema: Record<string, unknown>;
+  };
+  /** Thinking level for Claude models */
+  thinkingLevel?: ThinkingLevel;
+  /** If true, runs in read-only mode (no file writes) */
+  readOnly?: boolean;
+  /** Setting sources for CLAUDE.md loading */
+  settingSources?: Array<'user' | 'project' | 'local'>;
+}
+
+/**
+ * Result from a simple query
+ */
+export interface SimpleQueryResult {
+  /** The accumulated text response */
+  text: string;
+  /** Structured output if outputFormat was specified and provider supports it */
+  structured_output?: Record<string, unknown>;
+}
+
+/**
+ * Options for streaming query execution
+ */
+export interface StreamingQueryOptions extends SimpleQueryOptions {
+  /** Callback for each text chunk received */
+  onText?: (text: string) => void;
+  /** Callback for tool use events */
+  onToolUse?: (tool: string, input: unknown) => void;
+  /** Callback for thinking blocks (if available) */
+  onThinking?: (thinking: string) => void;
+}
+
+/**
+ * Default model to use when none specified
+ */
+const DEFAULT_MODEL = 'claude-sonnet-4-20250514';
+
+/**
+ * Execute a simple query and return the text result
+ *
+ * Use this for simple, non-streaming queries where you just need
+ * the final text response. For more complex use cases with progress
+ * callbacks, use streamingQuery() instead.
+ *
+ * @example
+ * ```typescript
+ * const result = await simpleQuery({
+ *   prompt: 'Generate a title for: user authentication',
+ *   cwd: process.cwd(),
+ *   systemPrompt: 'You are a title generator...',
+ *   maxTurns: 1,
+ *   allowedTools: [],
+ * });
+ * console.log(result.text); // "Add user authentication"
+ * ```
+ */
+export async function simpleQuery(options: SimpleQueryOptions): Promise<SimpleQueryResult> {
+  const model = options.model || DEFAULT_MODEL;
+  const provider = ProviderFactory.getProviderForModel(model);
+
+  let responseText = '';
+  let structuredOutput: Record<string, unknown> | undefined;
+
+  // Build provider options
+  const providerOptions = {
+    prompt: options.prompt,
+    model: model,
+    cwd: options.cwd,
+    systemPrompt: options.systemPrompt,
+    maxTurns: options.maxTurns ?? 1,
+    allowedTools: options.allowedTools ?? [],
+    abortController: options.abortController,
+    outputFormat: options.outputFormat,
+    thinkingLevel: options.thinkingLevel,
+    readOnly: options.readOnly,
+    settingSources: options.settingSources,
+  };
+
+  for await (const msg of provider.executeQuery(providerOptions)) {
+    // Handle error messages
+    if (msg.type === 'error') {
+      const errorMessage = msg.error || 'Provider returned an error';
+      throw new Error(errorMessage);
+    }
+
+    // Extract text from assistant messages
+    if (msg.type === 'assistant' && msg.message?.content) {
+      for (const block of msg.message.content) {
+        if (block.type === 'text' && block.text) {
+          responseText += block.text;
+        }
+      }
+    }
+
+    // Handle result messages
+    if (msg.type === 'result') {
+      if (msg.subtype === 'success') {
+        // Use result text if longer than accumulated text
+        if (msg.result && msg.result.length > responseText.length) {
+          responseText = msg.result;
+        }
+        // Capture structured output if present
+        if (msg.structured_output) {
+          structuredOutput = msg.structured_output;
+        }
+      } else if (msg.subtype === 'error_max_turns') {
+        // Max turns reached - return what we have
+        break;
+      } else if (msg.subtype === 'error_max_structured_output_retries') {
+        throw new Error('Could not produce valid structured output after retries');
+      }
+    }
+  }
+
+  return { text: responseText, structured_output: structuredOutput };
+}
+
+/**
+ * Execute a streaming query with event callbacks
+ *
+ * Use this for queries where you need real-time progress updates,
+ * such as when displaying streaming output to a user.
+ *
+ * @example
+ * ```typescript
+ * const result = await streamingQuery({
+ *   prompt: 'Analyze this project and suggest improvements',
+ *   cwd: '/path/to/project',
+ *   maxTurns: 250,
+ *   allowedTools: ['Read', 'Glob', 'Grep'],
+ *   onText: (text) => emitProgress(text),
+ *   onToolUse: (tool, input) => emitToolUse(tool, input),
+ * });
+ * ```
+ */
+export async function streamingQuery(options: StreamingQueryOptions): Promise<SimpleQueryResult> {
+  const model = options.model || DEFAULT_MODEL;
+  const provider = ProviderFactory.getProviderForModel(model);
+
+  let responseText = '';
+  let structuredOutput: Record<string, unknown> | undefined;
+
+  // Build provider options
+  const providerOptions = {
+    prompt: options.prompt,
+    model: model,
+    cwd: options.cwd,
+    systemPrompt: options.systemPrompt,
+    maxTurns: options.maxTurns ?? 250,
+    allowedTools: options.allowedTools ?? ['Read', 'Glob', 'Grep'],
+    abortController: options.abortController,
+    outputFormat: options.outputFormat,
+    thinkingLevel: options.thinkingLevel,
+    readOnly: options.readOnly,
+    settingSources: options.settingSources,
+  };
+
+  for await (const msg of provider.executeQuery(providerOptions)) {
+    // Handle error messages
+    if (msg.type === 'error') {
+      const errorMessage = msg.error || 'Provider returned an error';
+      throw new Error(errorMessage);
+    }
+
+    // Extract content from assistant messages
+    if (msg.type === 'assistant' && msg.message?.content) {
+      for (const block of msg.message.content) {
+        if (block.type === 'text' && block.text) {
+          responseText += block.text;
+          options.onText?.(block.text);
+        } else if (block.type === 'tool_use' && block.name) {
+          options.onToolUse?.(block.name, block.input);
+        } else if (block.type === 'thinking' && block.thinking) {
+          options.onThinking?.(block.thinking);
+        }
+      }
+    }
+
+    // Handle result messages
+    if (msg.type === 'result') {
+      if (msg.subtype === 'success') {
+        // Use result text if longer than accumulated text
+        if (msg.result && msg.result.length > responseText.length) {
+          responseText = msg.result;
+        }
+        // Capture structured output if present
+        if (msg.structured_output) {
+          structuredOutput = msg.structured_output;
+        }
+      } else if (msg.subtype === 'error_max_turns') {
+        // Max turns reached - return what we have
+        break;
+      } else if (msg.subtype === 'error_max_structured_output_retries') {
+        throw new Error('Could not produce valid structured output after retries');
+      }
+    }
+  }
+
+  return { text: responseText, structured_output: structuredOutput };
+}

apps/server/src/providers/types.ts

@@ -2,6 +2,7 @@
  * Shared types for AI model providers
  *
  * Re-exports types from @automaker/types for consistency across the codebase.
+ * All provider types are defined in @automaker/types to avoid duplication.
  */
 
 // Re-export all provider types from @automaker/types
@@ -13,72 +14,9 @@ export type {
   McpStdioServerConfig,
   McpSSEServerConfig,
   McpHttpServerConfig,
+  ContentBlock,
+  ProviderMessage,
+  InstallationStatus,
+  ValidationResult,
+  ModelDefinition,
 } from '@automaker/types';
-
-/**
- * Content block in a provider message (matches Claude SDK format)
- */
-export interface ContentBlock {
-  type: 'text' | 'tool_use' | 'thinking' | 'tool_result';
-  text?: string;
-  thinking?: string;
-  name?: string;
-  input?: unknown;
-  tool_use_id?: string;
-  content?: string;
-}
-
-/**
- * Message returned by a provider (matches Claude SDK streaming format)
- */
-export interface ProviderMessage {
-  type: 'assistant' | 'user' | 'error' | 'result';
-  subtype?: 'success' | 'error';
-  session_id?: string;
-  message?: {
-    role: 'user' | 'assistant';
-    content: ContentBlock[];
-  };
-  result?: string;
-  error?: string;
-  parent_tool_use_id?: string | null;
-}
-
-/**
- * Installation status for a provider
- */
-export interface InstallationStatus {
-  installed: boolean;
-  path?: string;
-  version?: string;
-  method?: 'cli' | 'npm' | 'brew' | 'sdk';
-  hasApiKey?: boolean;
-  authenticated?: boolean;
-  error?: string;
-}
-
-/**
- * Validation result
- */
-export interface ValidationResult {
-  valid: boolean;
-  errors: string[];
-  warnings?: string[];
-}
-
-/**
- * Model definition
- */
-export interface ModelDefinition {
-  id: string;
-  name: string;
-  modelString: string;
-  provider: string;
-  description: string;
-  contextWindow?: number;
-  maxOutputTokens?: number;
-  supportsVision?: boolean;
-  supportsTools?: boolean;
-  tier?: 'basic' | 'standard' | 'premium';
-  default?: boolean;
-}

apps/server/src/routes/agent/routes/queue-add.ts

@@ -3,17 +3,19 @@
  */
 
 import type { Request, Response } from 'express';
+import type { ThinkingLevel } from '@automaker/types';
 import { AgentService } from '../../../services/agent-service.js';
 import { getErrorMessage, logError } from '../common.js';
 
 export function createQueueAddHandler(agentService: AgentService) {
   return async (req: Request, res: Response): Promise<void> => {
     try {
-      const { sessionId, message, imagePaths, model } = req.body as {
+      const { sessionId, message, imagePaths, model, thinkingLevel } = req.body as {
         sessionId: string;
         message: string;
         imagePaths?: string[];
         model?: string;
+        thinkingLevel?: ThinkingLevel;
       };
 
       if (!sessionId || !message) {
@@ -24,7 +26,12 @@ export function createQueueAddHandler(agentService: AgentService) {
         return;
       }
 
-      const result = await agentService.addToQueue(sessionId, { message, imagePaths, model });
+      const result = await agentService.addToQueue(sessionId, {
+        message,
+        imagePaths,
+        model,
+        thinkingLevel,
+      });
       res.json(result);
     } catch (error) {
       logError(error, 'Add to queue failed');

apps/server/src/routes/agent/routes/send.ts

@@ -3,6 +3,7 @@
  */
 
 import type { Request, Response } from 'express';
+import type { ThinkingLevel } from '@automaker/types';
 import { AgentService } from '../../../services/agent-service.js';
 import { createLogger } from '@automaker/utils';
 import { getErrorMessage, logError } from '../common.js';
@@ -11,24 +12,27 @@ const logger = createLogger('Agent');
 export function createSendHandler(agentService: AgentService) {
   return async (req: Request, res: Response): Promise<void> => {
     try {
-      const { sessionId, message, workingDirectory, imagePaths, model } = req.body as {
-        sessionId: string;
-        message: string;
-        workingDirectory?: string;
-        imagePaths?: string[];
-        model?: string;
-      };
+      const { sessionId, message, workingDirectory, imagePaths, model, thinkingLevel } =
+        req.body as {
+          sessionId: string;
+          message: string;
+          workingDirectory?: string;
+          imagePaths?: string[];
+          model?: string;
+          thinkingLevel?: ThinkingLevel;
+        };
 
-      console.log('[Send Handler] Received request:', {
+      logger.debug('Received request:', {
         sessionId,
         messageLength: message?.length,
         workingDirectory,
         imageCount: imagePaths?.length || 0,
         model,
+        thinkingLevel,
       });
 
       if (!sessionId || !message) {
-        console.log('[Send Handler] ERROR: Validation failed - missing sessionId or message');
+        logger.warn('Validation failed - missing sessionId or message');
         res.status(400).json({
           success: false,
           error: 'sessionId and message are required',
@@ -36,7 +40,7 @@ export function createSendHandler(agentService: AgentService) {
         return;
       }
 
-      console.log('[Send Handler] Validation passed, calling agentService.sendMessage()');
+      logger.debug('Validation passed, calling agentService.sendMessage()');
 
       // Start the message processing (don't await - it streams via WebSocket)
       agentService
@@ -46,18 +50,19 @@ export function createSendHandler(agentService: AgentService) {
           workingDirectory,
           imagePaths,
           model,
+          thinkingLevel,
         })
         .catch((error) => {
-          console.error('[Send Handler] ERROR: Background error in sendMessage():', error);
+          logger.error('Background error in sendMessage():', error);
           logError(error, 'Send message failed (background)');
         });
 
-      console.log('[Send Handler] Returning immediate response to client');
+      logger.debug('Returning immediate response to client');
 
       // Return immediately - responses come via WebSocket
       res.json({ success: true, message: 'Message sent' });
     } catch (error) {
-      console.error('[Send Handler] ERROR: Synchronous error:', error);
+      logger.error('Synchronous error:', error);
       logError(error, 'Send message failed');
       res.status(500).json({ success: false, error: getErrorMessage(error) });
     }

apps/server/src/routes/app-spec/common.ts

@@ -6,26 +6,57 @@ import { createLogger } from '@automaker/utils';
 
 const logger = createLogger('SpecRegeneration');
 
-// Shared state for tracking generation status - private
-let isRunning = false;
-let currentAbortController: AbortController | null = null;
+// Shared state for tracking generation status - scoped by project path
+const runningProjects = new Map<string, boolean>();
+const abortControllers = new Map<string, AbortController>();
 
 /**
- * Get the current running state
+ * Get the running state for a specific project
  */
-export function getSpecRegenerationStatus(): {
+export function getSpecRegenerationStatus(projectPath?: string): {
   isRunning: boolean;
   currentAbortController: AbortController | null;
+  projectPath?: string;
 } {
-  return { isRunning, currentAbortController };
+  if (projectPath) {
+    return {
+      isRunning: runningProjects.get(projectPath) || false,
+      currentAbortController: abortControllers.get(projectPath) || null,
+      projectPath,
+    };
+  }
+  // Fallback: check if any project is running (for backward compatibility)
+  const isAnyRunning = Array.from(runningProjects.values()).some((running) => running);
+  return { isRunning: isAnyRunning, currentAbortController: null };
 }
 
 /**
- * Set the running state and abort controller
+ * Get the project path that is currently running (if any)
  */
-export function setRunningState(running: boolean, controller: AbortController | null = null): void {
-  isRunning = running;
-  currentAbortController = controller;
+export function getRunningProjectPath(): string | null {
+  for (const [path, running] of runningProjects.entries()) {
+    if (running) return path;
+  }
+  return null;
+}
+
+/**
+ * Set the running state and abort controller for a specific project
+ */
+export function setRunningState(
+  projectPath: string,
+  running: boolean,
+  controller: AbortController | null = null
+): void {
+  if (running) {
+    runningProjects.set(projectPath, true);
+    if (controller) {
+      abortControllers.set(projectPath, controller);
+    }
+  } else {
+    runningProjects.delete(projectPath);
+    abortControllers.delete(projectPath);
+  }
 }
 
 /**

apps/server/src/routes/app-spec/generate-features-from-spec.ts

@@ -1,13 +1,16 @@
 /**
  * Generate features from existing app_spec.txt
+ *
+ * Model is configurable via phaseModels.featureGenerationModel in settings
+ * (defaults to Sonnet for balanced speed and quality).
  */
 
-import { query } from '@anthropic-ai/claude-agent-sdk';
 import * as secureFs from '../../lib/secure-fs.js';
 import type { EventEmitter } from '../../lib/events.js';
 import { createLogger } from '@automaker/utils';
-import { createFeatureGenerationOptions } from '../../lib/sdk-options.js';
-import { logAuthStatus } from './common.js';
+import { DEFAULT_PHASE_MODELS } from '@automaker/types';
+import { resolvePhaseModel } from '@automaker/model-resolver';
+import { streamingQuery } from '../../providers/simple-query-service.js';
 import { parseAndCreateFeatures } from './parse-and-create-features.js';
 import { getAppSpecPath } from '@automaker/platform';
 import type { SettingsService } from '../../services/settings-service.js';
@@ -101,67 +104,38 @@ IMPORTANT: Do not ask for clarification. The specification is provided above. Ge
     '[FeatureGeneration]'
   );
 
-  const options = createFeatureGenerationOptions({
+  // Get model from phase settings
+  const settings = await settingsService?.getGlobalSettings();
+  const phaseModelEntry =
+    settings?.phaseModels?.featureGenerationModel || DEFAULT_PHASE_MODELS.featureGenerationModel;
+  const { model, thinkingLevel } = resolvePhaseModel(phaseModelEntry);
+
+  logger.info('Using model:', model);
+
+  // Use streamingQuery with event callbacks
+  const result = await streamingQuery({
+    prompt,
+    model,
     cwd: projectPath,
+    maxTurns: 250,
+    allowedTools: ['Read', 'Glob', 'Grep'],
     abortController,
-    autoLoadClaudeMd,
+    thinkingLevel,
+    readOnly: true, // Feature generation only reads code, doesn't write
+    settingSources: autoLoadClaudeMd ? ['user', 'project', 'local'] : undefined,
+    onText: (text) => {
+      logger.debug(`Feature text block received (${text.length} chars)`);
+      events.emit('spec-regeneration:event', {
+        type: 'spec_regeneration_progress',
+        content: text,
+        projectPath: projectPath,
+      });
+    },
   });
 
-  logger.debug('SDK Options:', JSON.stringify(options, null, 2));
-  logger.info('Calling Claude Agent SDK query() for features...');
+  const responseText = result.text;
 
-  logAuthStatus('Right before SDK query() for features');
-
-  let stream;
-  try {
-    stream = query({ prompt, options });
-    logger.debug('query() returned stream successfully');
-  } catch (queryError) {
-    logger.error('❌ query() threw an exception:');
-    logger.error('Error:', queryError);
-    throw queryError;
-  }
-
-  let responseText = '';
-  let messageCount = 0;
-
-  logger.debug('Starting to iterate over feature stream...');
-
-  try {
-    for await (const msg of stream) {
-      messageCount++;
-      logger.debug(
-        `Feature stream message #${messageCount}:`,
-        JSON.stringify({ type: msg.type, subtype: (msg as any).subtype }, null, 2)
-      );
-
-      if (msg.type === 'assistant' && msg.message.content) {
-        for (const block of msg.message.content) {
-          if (block.type === 'text') {
-            responseText += block.text;
-            logger.debug(`Feature text block received (${block.text.length} chars)`);
-            events.emit('spec-regeneration:event', {
-              type: 'spec_regeneration_progress',
-              content: block.text,
-              projectPath: projectPath,
-            });
-          }
-        }
-      } else if (msg.type === 'result' && (msg as any).subtype === 'success') {
-        logger.debug('Received success result for features');
-        responseText = (msg as any).result || responseText;
-      } else if ((msg as { type: string }).type === 'error') {
-        logger.error('❌ Received error message from feature stream:');
-        logger.error('Error message:', JSON.stringify(msg, null, 2));
-      }
-    }
-  } catch (streamError) {
-    logger.error('❌ Error while iterating feature stream:');
-    logger.error('Stream error:', streamError);
-    throw streamError;
-  }
-
-  logger.info(`Feature stream complete. Total messages: ${messageCount}`);
+  logger.info(`Feature stream complete.`);
   logger.info(`Feature response length: ${responseText.length} chars`);
   logger.info('========== FULL RESPONSE TEXT ==========');
   logger.info(responseText);

apps/server/src/routes/app-spec/generate-spec.ts

@@ -1,9 +1,10 @@
 /**
  * Generate app_spec.txt from project overview
+ *
+ * Model is configurable via phaseModels.specGenerationModel in settings
+ * (defaults to Opus for high-quality specification generation).
  */
 
-import { query } from '@anthropic-ai/claude-agent-sdk';
-import path from 'path';
 import * as secureFs from '../../lib/secure-fs.js';
 import type { EventEmitter } from '../../lib/events.js';
 import {
@@ -13,8 +14,10 @@ import {
   type SpecOutput,
 } from '../../lib/app-spec-format.js';
 import { createLogger } from '@automaker/utils';
-import { createSpecGenerationOptions } from '../../lib/sdk-options.js';
-import { logAuthStatus } from './common.js';
+import { DEFAULT_PHASE_MODELS, isCursorModel } from '@automaker/types';
+import { resolvePhaseModel } from '@automaker/model-resolver';
+import { extractJson } from '../../lib/json-extractor.js';
+import { streamingQuery } from '../../providers/simple-query-service.js';
 import { generateFeaturesFromSpec } from './generate-features-from-spec.js';
 import { ensureAutomakerDir, getAppSpecPath } from '@automaker/platform';
 import type { SettingsService } from '../../services/settings-service.js';
@@ -93,105 +96,84 @@ ${getStructuredSpecPromptInstruction()}`;
     '[SpecRegeneration]'
   );
 
-  const options = createSpecGenerationOptions({
+  // Get model from phase settings
+  const settings = await settingsService?.getGlobalSettings();
+  const phaseModelEntry =
+    settings?.phaseModels?.specGenerationModel || DEFAULT_PHASE_MODELS.specGenerationModel;
+  const { model, thinkingLevel } = resolvePhaseModel(phaseModelEntry);
+
+  logger.info('Using model:', model);
+
+  let responseText = '';
+  let structuredOutput: SpecOutput | null = null;
+
+  // Determine if we should use structured output (Claude supports it, Cursor doesn't)
+  const useStructuredOutput = !isCursorModel(model);
+
+  // Build the final prompt - for Cursor, include JSON schema instructions
+  let finalPrompt = prompt;
+  if (!useStructuredOutput) {
+    finalPrompt = `${prompt}
+
+CRITICAL INSTRUCTIONS:
+1. DO NOT write any files. DO NOT create any files like "project_specification.json".
+2. After analyzing the project, respond with ONLY a JSON object - no explanations, no markdown, just raw JSON.
+3. The JSON must match this exact schema:
+
+${JSON.stringify(specOutputSchema, null, 2)}
+
+Your entire response should be valid JSON starting with { and ending with }. No text before or after.`;
+  }
+
+  // Use streamingQuery with event callbacks
+  const result = await streamingQuery({
+    prompt: finalPrompt,
+    model,
     cwd: projectPath,
+    maxTurns: 250,
+    allowedTools: ['Read', 'Glob', 'Grep'],
     abortController,
-    autoLoadClaudeMd,
-    outputFormat: {
-      type: 'json_schema',
-      schema: specOutputSchema,
+    thinkingLevel,
+    readOnly: true, // Spec generation only reads code, we write the spec ourselves
+    settingSources: autoLoadClaudeMd ? ['user', 'project', 'local'] : undefined,
+    outputFormat: useStructuredOutput
+      ? {
+          type: 'json_schema',
+          schema: specOutputSchema,
+        }
+      : undefined,
+    onText: (text) => {
+      responseText += text;
+      logger.info(
+        `Text block received (${text.length} chars), total now: ${responseText.length} chars`
+      );
+      events.emit('spec-regeneration:event', {
+        type: 'spec_regeneration_progress',
+        content: text,
+        projectPath: projectPath,
+      });
+    },
+    onToolUse: (tool, input) => {
+      logger.info('Tool use:', tool);
+      events.emit('spec-regeneration:event', {
+        type: 'spec_tool',
+        tool,
+        input,
+      });
     },
   });
 
-  logger.debug('SDK Options:', JSON.stringify(options, null, 2));
-  logger.info('Calling Claude Agent SDK query()...');
-
-  // Log auth status right before the SDK call
-  logAuthStatus('Right before SDK query()');
-
-  let stream;
-  try {
-    stream = query({ prompt, options });
-    logger.debug('query() returned stream successfully');
-  } catch (queryError) {
-    logger.error('❌ query() threw an exception:');
-    logger.error('Error:', queryError);
-    throw queryError;
+  // Get structured output if available
+  if (result.structured_output) {
+    structuredOutput = result.structured_output as unknown as SpecOutput;
+    logger.info('✅ Received structured output');
+    logger.debug('Structured output:', JSON.stringify(structuredOutput, null, 2));
+  } else if (!useStructuredOutput && responseText) {
+    // For non-Claude providers, parse JSON from response text
+    structuredOutput = extractJson<SpecOutput>(responseText, { logger });
   }
 
-  let responseText = '';
-  let messageCount = 0;
-  let structuredOutput: SpecOutput | null = null;
-
-  logger.info('Starting to iterate over stream...');
-
-  try {
-    for await (const msg of stream) {
-      messageCount++;
-      logger.info(
-        `Stream message #${messageCount}: type=${msg.type}, subtype=${(msg as any).subtype}`
-      );
-
-      if (msg.type === 'assistant') {
-        const msgAny = msg as any;
-        if (msgAny.message?.content) {
-          for (const block of msgAny.message.content) {
-            if (block.type === 'text') {
-              responseText += block.text;
-              logger.info(
-                `Text block received (${block.text.length} chars), total now: ${responseText.length} chars`
-              );
-              events.emit('spec-regeneration:event', {
-                type: 'spec_regeneration_progress',
-                content: block.text,
-                projectPath: projectPath,
-              });
-            } else if (block.type === 'tool_use') {
-              logger.info('Tool use:', block.name);
-              events.emit('spec-regeneration:event', {
-                type: 'spec_tool',
-                tool: block.name,
-                input: block.input,
-              });
-            }
-          }
-        }
-      } else if (msg.type === 'result' && (msg as any).subtype === 'success') {
-        logger.info('Received success result');
-        // Check for structured output - this is the reliable way to get spec data
-        const resultMsg = msg as any;
-        if (resultMsg.structured_output) {
-          structuredOutput = resultMsg.structured_output as SpecOutput;
-          logger.info('✅ Received structured output');
-          logger.debug('Structured output:', JSON.stringify(structuredOutput, null, 2));
-        } else {
-          logger.warn('⚠️ No structured output in result, will fall back to text parsing');
-        }
-      } else if (msg.type === 'result') {
-        // Handle error result types
-        const subtype = (msg as any).subtype;
-        logger.info(`Result message: subtype=${subtype}`);
-        if (subtype === 'error_max_turns') {
-          logger.error('❌ Hit max turns limit!');
-        } else if (subtype === 'error_max_structured_output_retries') {
-          logger.error('❌ Failed to produce valid structured output after retries');
-          throw new Error('Could not produce valid spec output');
-        }
-      } else if ((msg as { type: string }).type === 'error') {
-        logger.error('❌ Received error message from stream:');
-        logger.error('Error message:', JSON.stringify(msg, null, 2));
-      } else if (msg.type === 'user') {
-        // Log user messages (tool results)
-        logger.info(`User message (tool result): ${JSON.stringify(msg).substring(0, 500)}`);
-      }
-    }
-  } catch (streamError) {
-    logger.error('❌ Error while iterating stream:');
-    logger.error('Stream error:', streamError);
-    throw streamError;
-  }
-
-  logger.info(`Stream iteration complete. Total messages: ${messageCount}`);
+  logger.info(`Stream iteration complete.`);
   logger.info(`Response text length: ${responseText.length} chars`);
 
   // Determine XML content to save

apps/server/src/routes/app-spec/parse-and-create-features.ts

@@ -7,6 +7,7 @@ import * as secureFs from '../../lib/secure-fs.js';
 import type { EventEmitter } from '../../lib/events.js';
 import { createLogger } from '@automaker/utils';
 import { getFeaturesDir } from '@automaker/platform';
+import { extractJsonWithArray } from '../../lib/json-extractor.js';
 
 const logger = createLogger('SpecRegeneration');
 
@@ -22,23 +23,30 @@ export async function parseAndCreateFeatures(
   logger.info('========== END CONTENT ==========');
 
   try {
-    // Extract JSON from response
-    logger.info('Extracting JSON from response...');
-    logger.info(`Looking for pattern: /{[\\s\\S]*"features"[\\s\\S]*}/`);
-    const jsonMatch = content.match(/\{[\s\S]*"features"[\s\S]*\}/);
-    if (!jsonMatch) {
-      logger.error('❌ No valid JSON found in response');
+    // Extract JSON from response using shared utility
+    logger.info('Extracting JSON from response using extractJsonWithArray...');
+
+    interface FeaturesResponse {
+      features: Array<{
+        id: string;
+        category?: string;
+        title: string;
+        description: string;
+        priority?: number;
+        complexity?: string;
+        dependencies?: string[];
+      }>;
+    }
+
+    const parsed = extractJsonWithArray<FeaturesResponse>(content, 'features', { logger });
+
+    if (!parsed || !parsed.features) {
+      logger.error('❌ No valid JSON with "features" array found in response');
       logger.error('Full content received:');
       logger.error(content);
       throw new Error('No valid JSON found in response');
     }
 
-    logger.info(`JSON match found (${jsonMatch[0].length} chars)`);
-    logger.info('========== MATCHED JSON ==========');
-    logger.info(jsonMatch[0]);
-    logger.info('========== END MATCHED JSON ==========');
-
-    const parsed = JSON.parse(jsonMatch[0]);
     logger.info(`Parsed ${parsed.features?.length || 0} features`);
     logger.info('Parsed features:', JSON.stringify(parsed.features, null, 2));
 

apps/server/src/routes/app-spec/routes/create.ts

@@ -47,17 +47,17 @@ export function createCreateHandler(events: EventEmitter) {
         return;
       }
 
-      const { isRunning } = getSpecRegenerationStatus();
+      const { isRunning } = getSpecRegenerationStatus(projectPath);
       if (isRunning) {
-        logger.warn('Generation already running, rejecting request');
-        res.json({ success: false, error: 'Spec generation already running' });
+        logger.warn('Generation already running for project:', projectPath);
+        res.json({ success: false, error: 'Spec generation already running for this project' });
         return;
       }
 
       logAuthStatus('Before starting generation');
 
       const abortController = new AbortController();
-      setRunningState(true, abortController);
+      setRunningState(projectPath, true, abortController);
       logger.info('Starting background generation task...');
 
       // Start generation in background
@@ -80,7 +80,7 @@ export function createCreateHandler(events: EventEmitter) {
         })
         .finally(() => {
           logger.info('Generation task finished (success or error)');
-          setRunningState(false, null);
+          setRunningState(projectPath, false, null);
         });
 
       logger.info('Returning success response (generation running in background)');

apps/server/src/routes/app-spec/routes/generate-features.ts

@@ -40,17 +40,17 @@ export function createGenerateFeaturesHandler(
         return;
       }
 
-      const { isRunning } = getSpecRegenerationStatus();
+      const { isRunning } = getSpecRegenerationStatus(projectPath);
       if (isRunning) {
-        logger.warn('Generation already running, rejecting request');
-        res.json({ success: false, error: 'Generation already running' });
+        logger.warn('Generation already running for project:', projectPath);
+        res.json({ success: false, error: 'Generation already running for this project' });
         return;
       }
 
       logAuthStatus('Before starting feature generation');
 
       const abortController = new AbortController();
-      setRunningState(true, abortController);
+      setRunningState(projectPath, true, abortController);
       logger.info('Starting background feature generation task...');
 
       generateFeaturesFromSpec(projectPath, events, abortController, maxFeatures, settingsService)
@@ -63,7 +63,7 @@ export function createGenerateFeaturesHandler(
         })
         .finally(() => {
           logger.info('Feature generation task finished (success or error)');
-          setRunningState(false, null);
+          setRunningState(projectPath, false, null);
         });
 
       logger.info('Returning success response (generation running in background)');

apps/server/src/routes/app-spec/routes/generate.ts

@@ -48,17 +48,17 @@ export function createGenerateHandler(events: EventEmitter, settingsService?: Se
         return;
       }
 
-      const { isRunning } = getSpecRegenerationStatus();
+      const { isRunning } = getSpecRegenerationStatus(projectPath);
       if (isRunning) {
-        logger.warn('Generation already running, rejecting request');
-        res.json({ success: false, error: 'Spec generation already running' });
+        logger.warn('Generation already running for project:', projectPath);
+        res.json({ success: false, error: 'Spec generation already running for this project' });
         return;
       }
 
       logAuthStatus('Before starting generation');
 
       const abortController = new AbortController();
-      setRunningState(true, abortController);
+      setRunningState(projectPath, true, abortController);
       logger.info('Starting background generation task...');
 
       generateSpec(
@@ -81,7 +81,7 @@ export function createGenerateHandler(events: EventEmitter, settingsService?: Se
         })
         .finally(() => {
           logger.info('Generation task finished (success or error)');
-          setRunningState(false, null);
+          setRunningState(projectPath, false, null);
         });
 
       logger.info('Returning success response (generation running in background)');

apps/server/src/routes/app-spec/routes/status.ts

@@ -6,10 +6,11 @@ import type { Request, Response } from 'express';
 import { getSpecRegenerationStatus, getErrorMessage } from '../common.js';
 
 export function createStatusHandler() {
-  return async (_req: Request, res: Response): Promise<void> => {
+  return async (req: Request, res: Response): Promise<void> => {
     try {
-      const { isRunning } = getSpecRegenerationStatus();
-      res.json({ success: true, isRunning });
+      const projectPath = req.query.projectPath as string | undefined;
+      const { isRunning } = getSpecRegenerationStatus(projectPath);
+      res.json({ success: true, isRunning, projectPath });
     } catch (error) {
       res.status(500).json({ success: false, error: getErrorMessage(error) });
     }

apps/server/src/routes/app-spec/routes/stop.ts

@@ -6,13 +6,16 @@ import type { Request, Response } from 'express';
 import { getSpecRegenerationStatus, setRunningState, getErrorMessage } from '../common.js';
 
 export function createStopHandler() {
-  return async (_req: Request, res: Response): Promise<void> => {
+  return async (req: Request, res: Response): Promise<void> => {
     try {
-      const { currentAbortController } = getSpecRegenerationStatus();
+      const { projectPath } = req.body as { projectPath?: string };
+      const { currentAbortController } = getSpecRegenerationStatus(projectPath);
       if (currentAbortController) {
         currentAbortController.abort();
       }
-      setRunningState(false, null);
+      if (projectPath) {
+        setRunningState(projectPath, false, null);
+      }
       res.json({ success: true });
     } catch (error) {
       res.status(500).json({ success: false, error: getErrorMessage(error) });

apps/server/src/routes/auth/index.ts

@@ -229,12 +229,13 @@ export function createAuthRoutes(): Router {
       await invalidateSession(sessionToken);
     }
 
-    // Clear the cookie
-    res.clearCookie(cookieName, {
-      httpOnly: true,
-      secure: process.env.NODE_ENV === 'production',
-      sameSite: 'strict',
-      path: '/',
+    // Clear the cookie by setting it to empty with immediate expiration
+    // Using res.cookie() with maxAge: 0 is more reliable than clearCookie()
+    // in cross-origin development environments
+    res.cookie(cookieName, '', {
+      ...getSessionCookieOptions(),
+      maxAge: 0,
+      expires: new Date(0),
     });
 
     res.json({

apps/server/src/routes/auto-mode/index.ts

@@ -17,6 +17,7 @@ import { createAnalyzeProjectHandler } from './routes/analyze-project.js';
 import { createFollowUpFeatureHandler } from './routes/follow-up-feature.js';
 import { createCommitFeatureHandler } from './routes/commit-feature.js';
 import { createApprovePlanHandler } from './routes/approve-plan.js';
+import { createResumeInterruptedHandler } from './routes/resume-interrupted.js';
 
 export function createAutoModeRoutes(autoModeService: AutoModeService): Router {
   const router = Router();
@@ -63,6 +64,11 @@ export function createAutoModeRoutes(autoModeService: AutoModeService): Router {
     validatePathParams('projectPath'),
     createApprovePlanHandler(autoModeService)
   );
+  router.post(
+    '/resume-interrupted',
+    validatePathParams('projectPath'),
+    createResumeInterruptedHandler(autoModeService)
+  );
 
   return router;
 }

apps/server/src/routes/auto-mode/routes/follow-up-feature.ts

@@ -31,7 +31,9 @@ export function createFollowUpFeatureHandler(autoModeService: AutoModeService) {
       // Start follow-up in background
       // followUpFeature derives workDir from feature.branchName
       autoModeService
-        .followUpFeature(projectPath, featureId, prompt, imagePaths, useWorktrees ?? true)
+        // Default to false to match run-feature/resume-feature behavior.
+        // Worktrees should only be used when explicitly enabled by the user.
+        .followUpFeature(projectPath, featureId, prompt, imagePaths, useWorktrees ?? false)
         .catch((error) => {
           logger.error(`[AutoMode] Follow up feature ${featureId} error:`, error);
         })

apps/server/src/routes/auto-mode/routes/resume-feature.ts

@@ -31,7 +31,7 @@ export function createResumeFeatureHandler(autoModeService: AutoModeService) {
       autoModeService
         .resumeFeature(projectPath, featureId, useWorktrees ?? false)
         .catch((error) => {
-          logger.error(`[AutoMode] Resume feature ${featureId} error:`, error);
+          logger.error(`Resume feature ${featureId} error:`, error);
         });
 
       res.json({ success: true });

apps/server/src/routes/auto-mode/routes/resume-interrupted.ts

@@ -0,0 +1,42 @@
+/**
+ * Resume Interrupted Features Handler
+ *
+ * Checks for features that were interrupted (in pipeline steps or in_progress)
+ * when the server was restarted and resumes them.
+ */
+
+import type { Request, Response } from 'express';
+import { createLogger } from '@automaker/utils';
+import type { AutoModeService } from '../../../services/auto-mode-service.js';
+
+const logger = createLogger('ResumeInterrupted');
+
+interface ResumeInterruptedRequest {
+  projectPath: string;
+}
+
+export function createResumeInterruptedHandler(autoModeService: AutoModeService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    const { projectPath } = req.body as ResumeInterruptedRequest;
+
+    if (!projectPath) {
+      res.status(400).json({ error: 'Project path is required' });
+      return;
+    }
+
+    logger.info(`Checking for interrupted features in ${projectPath}`);
+
+    try {
+      await autoModeService.resumeInterruptedFeatures(projectPath);
+      res.json({
+        success: true,
+        message: 'Resume check completed',
+      });
+    } catch (error) {
+      logger.error('Error resuming interrupted features:', error);
+      res.status(500).json({
+        error: error instanceof Error ? error.message : 'Unknown error',
+      });
+    }
+  };
+}

apps/server/src/routes/auto-mode/routes/run-feature.ts

@@ -31,7 +31,7 @@ export function createRunFeatureHandler(autoModeService: AutoModeService) {
       autoModeService
         .executeFeature(projectPath, featureId, useWorktrees ?? false, false)
         .catch((error) => {
-          logger.error(`[AutoMode] Feature ${featureId} error:`, error);
+          logger.error(`Feature ${featureId} error:`, error);
         })
         .finally(() => {
           // Release the starting slot when execution completes (success or error)

apps/server/src/routes/backlog-plan/generate-plan.ts

@@ -1,11 +1,22 @@
 /**
  * Generate backlog plan using Claude AI
+ *
+ * Model is configurable via phaseModels.backlogPlanningModel in settings
+ * (defaults to Sonnet). Can be overridden per-call via model parameter.
  */
 
 import type { EventEmitter } from '../../lib/events.js';
 import type { Feature, BacklogPlanResult, BacklogChange, DependencyUpdate } from '@automaker/types';
+import {
+  DEFAULT_PHASE_MODELS,
+  isCursorModel,
+  stripProviderPrefix,
+  type ThinkingLevel,
+} from '@automaker/types';
+import { resolvePhaseModel } from '@automaker/model-resolver';
 import { FeatureLoader } from '../../services/feature-loader.js';
 import { ProviderFactory } from '../../providers/provider-factory.js';
+import { extractJsonWithArray } from '../../lib/json-extractor.js';
 import { logger, setRunningState, getErrorMessage } from './common.js';
 import type { SettingsService } from '../../services/settings-service.js';
 import { getAutoLoadClaudeMdSetting, getPromptCustomization } from '../../lib/settings-helpers.js';
@@ -39,24 +50,28 @@ function formatFeaturesForPrompt(features: Feature[]): string {
  * Parse the AI response into a BacklogPlanResult
  */
 function parsePlanResponse(response: string): BacklogPlanResult {
-  try {
-    // Try to extract JSON from the response
-    const jsonMatch = response.match(/```json\n?([\s\S]*?)\n?```/);
-    if (jsonMatch) {
-      return JSON.parse(jsonMatch[1]);
-    }
+  // Use shared JSON extraction utility for robust parsing
+  // extractJsonWithArray validates that 'changes' exists AND is an array
+  const parsed = extractJsonWithArray<BacklogPlanResult>(response, 'changes', {
+    logger,
+  });
 
-    // Try to parse the whole response as JSON
-    return JSON.parse(response);
-  } catch {
-    // If parsing fails, return an empty result
-    logger.warn('[BacklogPlan] Failed to parse AI response as JSON');
-    return {
-      changes: [],
-      summary: 'Failed to parse AI response',
-      dependencyUpdates: [],
-    };
+  if (parsed) {
+    return parsed;
   }
+
+  // If parsing fails, log details and return an empty result
+  logger.warn('[BacklogPlan] Failed to parse AI response as JSON');
+  logger.warn('[BacklogPlan] Response text length:', response.length);
+  logger.warn('[BacklogPlan] Response preview:', response.slice(0, 500));
+  if (response.length === 0) {
+    logger.error('[BacklogPlan] Response text is EMPTY! No content was extracted from stream.');
+  }
+  return {
+    changes: [],
+    summary: 'Failed to parse AI response',
+    dependencyUpdates: [],
+  };
 }
 
 /**
@@ -96,9 +111,22 @@ export async function generateBacklogPlan(
       content: 'Generating plan with AI...',
     });
 
-    // Get the model to use
-    const effectiveModel = model || 'sonnet';
+    // Get the model to use from settings or provided override
+    let effectiveModel = model;
+    let thinkingLevel: ThinkingLevel | undefined;
+    if (!effectiveModel) {
+      const settings = await settingsService?.getGlobalSettings();
+      const phaseModelEntry =
+        settings?.phaseModels?.backlogPlanningModel || DEFAULT_PHASE_MODELS.backlogPlanningModel;
+      const resolved = resolvePhaseModel(phaseModelEntry);
+      effectiveModel = resolved.model;
+      thinkingLevel = resolved.thinkingLevel;
+    }
+    logger.info('[BacklogPlan] Using model:', effectiveModel);
+
     const provider = ProviderFactory.getProviderForModel(effectiveModel);
+    // Strip provider prefix - providers expect bare model IDs
+    const bareModel = stripProviderPrefix(effectiveModel);
 
     // Get autoLoadClaudeMd setting
     const autoLoadClaudeMd = await getAutoLoadClaudeMdSetting(
@@ -107,16 +135,38 @@ export async function generateBacklogPlan(
       '[BacklogPlan]'
     );
 
+    // For Cursor models, we need to combine prompts with explicit instructions
+    // because Cursor doesn't support systemPrompt separation like Claude SDK
+    let finalPrompt = userPrompt;
+    let finalSystemPrompt: string | undefined = systemPrompt;
+
+    if (isCursorModel(effectiveModel)) {
+      logger.info('[BacklogPlan] Using Cursor model - adding explicit no-file-write instructions');
+      finalPrompt = `${systemPrompt}
+
+CRITICAL INSTRUCTIONS:
+1. DO NOT write any files. Return the JSON in your response only.
+2. DO NOT use Write, Edit, or any file modification tools.
+3. Respond with ONLY a JSON object - no explanations, no markdown, just raw JSON.
+4. Your entire response should be valid JSON starting with { and ending with }.
+5. No text before or after the JSON object.
+
+${userPrompt}`;
+      finalSystemPrompt = undefined; // System prompt is now embedded in the user prompt
+    }
+
     // Execute the query
     const stream = provider.executeQuery({
-      prompt: userPrompt,
-      model: effectiveModel,
+      prompt: finalPrompt,
+      model: bareModel,
       cwd: projectPath,
-      systemPrompt,
+      systemPrompt: finalSystemPrompt,
       maxTurns: 1,
       allowedTools: [], // No tools needed for this
       abortController,
       settingSources: autoLoadClaudeMd ? ['user', 'project'] : undefined,
+      readOnly: true, // Plan generation only generates text, doesn't write files
+      thinkingLevel, // Pass thinking level for extended thinking
     });
 
     let responseText = '';
@@ -134,6 +184,16 @@ export async function generateBacklogPlan(
             }
           }
         }
+      } else if (msg.type === 'result' && msg.subtype === 'success' && msg.result) {
+        // Use result if it's a final accumulated message (from Cursor provider)
+        logger.info('[BacklogPlan] Received result from Cursor, length:', msg.result.length);
+        logger.info('[BacklogPlan] Previous responseText length:', responseText.length);
+        if (msg.result.length > responseText.length) {
+          logger.info('[BacklogPlan] Using Cursor result (longer than accumulated text)');
+          responseText = msg.result;
+        } else {
+          logger.info('[BacklogPlan] Keeping accumulated text (longer than Cursor result)');
+        }
       }
     }
 

apps/server/src/routes/backlog-plan/routes/apply.ts

@@ -12,11 +12,22 @@ const featureLoader = new FeatureLoader();
 export function createApplyHandler() {
   return async (req: Request, res: Response): Promise<void> => {
     try {
-      const { projectPath, plan } = req.body as {
+      const {
+        projectPath,
+        plan,
+        branchName: rawBranchName,
+      } = req.body as {
         projectPath: string;
         plan: BacklogPlanResult;
+        branchName?: string;
       };
 
+      // Validate branchName: must be undefined or a non-empty trimmed string
+      const branchName =
+        typeof rawBranchName === 'string' && rawBranchName.trim().length > 0
+          ? rawBranchName.trim()
+          : undefined;
+
       if (!projectPath) {
         res.status(400).json({ success: false, error: 'projectPath required' });
         return;
@@ -82,6 +93,7 @@ export function createApplyHandler() {
             dependencies: change.feature.dependencies,
             priority: change.feature.priority,
             status: 'backlog',
+            branchName,
           });
 
           appliedChanges.push(`added:${newFeature.id}`);

apps/server/src/routes/claude/index.ts

@@ -1,5 +1,8 @@
 import { Router, Request, Response } from 'express';
 import { ClaudeUsageService } from '../../services/claude-usage-service.js';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('Claude');
 
 export function createClaudeRoutes(service: ClaudeUsageService): Router {
   const router = Router();
@@ -10,7 +13,10 @@ export function createClaudeRoutes(service: ClaudeUsageService): Router {
       // Check if Claude CLI is available first
       const isAvailable = await service.isAvailable();
       if (!isAvailable) {
-        res.status(503).json({
+        // IMPORTANT: This endpoint is behind Automaker session auth already.
+        // Use a 200 + error payload for Claude CLI issues so the UI doesn't
+        // interpret it as an invalid Automaker session (401/403 triggers logout).
+        res.status(200).json({
           error: 'Claude CLI not found',
           message: "Please install Claude Code CLI and run 'claude login' to authenticate",
         });
@@ -23,17 +29,18 @@ export function createClaudeRoutes(service: ClaudeUsageService): Router {
       const message = error instanceof Error ? error.message : 'Unknown error';
 
       if (message.includes('Authentication required') || message.includes('token_expired')) {
-        res.status(401).json({
+        // Do NOT use 401/403 here: that status code is reserved for Automaker session auth.
+        res.status(200).json({
           error: 'Authentication required',
           message: "Please run 'claude login' to authenticate",
         });
       } else if (message.includes('timed out')) {
-        res.status(504).json({
+        res.status(200).json({
           error: 'Command timed out',
           message: 'The Claude CLI took too long to respond',
         });
       } else {
-        console.error('Error fetching usage:', error);
+        logger.error('Error fetching usage:', error);
         res.status(500).json({ error: message });
       }
     }

apps/server/src/routes/codex/index.ts

@@ -0,0 +1,90 @@
+import { Router, Request, Response } from 'express';
+import { CodexUsageService } from '../../services/codex-usage-service.js';
+import { CodexModelCacheService } from '../../services/codex-model-cache-service.js';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('Codex');
+
+export function createCodexRoutes(
+  usageService: CodexUsageService,
+  modelCacheService: CodexModelCacheService
+): Router {
+  const router = Router();
+
+  // Get current usage (attempts to fetch from Codex CLI)
+  router.get('/usage', async (_req: Request, res: Response) => {
+    try {
+      // Check if Codex CLI is available first
+      const isAvailable = await usageService.isAvailable();
+      if (!isAvailable) {
+        // IMPORTANT: This endpoint is behind Automaker session auth already.
+        // Use a 200 + error payload for Codex CLI issues so the UI doesn't
+        // interpret it as an invalid Automaker session (401/403 triggers logout).
+        res.status(200).json({
+          error: 'Codex CLI not found',
+          message: "Please install Codex CLI and run 'codex login' to authenticate",
+        });
+        return;
+      }
+
+      const usage = await usageService.fetchUsageData();
+      res.json(usage);
+    } catch (error) {
+      const message = error instanceof Error ? error.message : 'Unknown error';
+
+      if (message.includes('not authenticated') || message.includes('login')) {
+        // Do NOT use 401/403 here: that status code is reserved for Automaker session auth.
+        res.status(200).json({
+          error: 'Authentication required',
+          message: "Please run 'codex login' to authenticate",
+        });
+      } else if (message.includes('not available') || message.includes('does not provide')) {
+        // This is the expected case - Codex doesn't provide usage stats
+        res.status(200).json({
+          error: 'Usage statistics not available',
+          message: message,
+        });
+      } else if (message.includes('timed out')) {
+        res.status(200).json({
+          error: 'Command timed out',
+          message: 'The Codex CLI took too long to respond',
+        });
+      } else {
+        logger.error('Error fetching usage:', error);
+        res.status(500).json({ error: message });
+      }
+    }
+  });
+
+  // Get available Codex models (cached)
+  router.get('/models', async (req: Request, res: Response) => {
+    try {
+      const forceRefresh = req.query.refresh === 'true';
+      const { models, cachedAt } = await modelCacheService.getModelsWithMetadata(forceRefresh);
+
+      if (models.length === 0) {
+        res.status(503).json({
+          success: false,
+          error: 'Codex CLI not available or not authenticated',
+          message: "Please install Codex CLI and run 'codex login' to authenticate",
+        });
+        return;
+      }
+
+      res.json({
+        success: true,
+        models,
+        cachedAt,
+      });
+    } catch (error) {
+      logger.error('Error fetching models:', error);
+      const message = error instanceof Error ? error.message : 'Unknown error';
+      res.status(500).json({
+        success: false,
+        error: message,
+      });
+    }
+  });
+
+  return router;
+}

apps/server/src/routes/context/routes/describe-file.ts

@@ -1,8 +1,9 @@
 /**
  * POST /context/describe-file endpoint - Generate description for a text file
  *
- * Uses Claude Haiku to analyze a text file and generate a concise description
- * suitable for context file metadata.
+ * Uses AI to analyze a text file and generate a concise description
+ * suitable for context file metadata. Model is configurable via
+ * phaseModels.fileDescriptionModel in settings (defaults to Haiku).
  *
  * SECURITY: This endpoint validates file paths against ALLOWED_ROOT_DIRECTORY
  * and reads file content directly (not via Claude's Read tool) to prevent
@@ -10,11 +11,11 @@
  */
 
 import type { Request, Response } from 'express';
-import { query } from '@anthropic-ai/claude-agent-sdk';
 import { createLogger } from '@automaker/utils';
-import { CLAUDE_MODEL_MAP } from '@automaker/types';
+import { DEFAULT_PHASE_MODELS } from '@automaker/types';
 import { PathNotAllowedError } from '@automaker/platform';
-import { createCustomOptions } from '../../../lib/sdk-options.js';
+import { resolvePhaseModel } from '@automaker/model-resolver';
+import { simpleQuery } from '../../../providers/simple-query-service.js';
 import * as secureFs from '../../../lib/secure-fs.js';
 import * as path from 'path';
 import type { SettingsService } from '../../../services/settings-service.js';
@@ -46,31 +47,6 @@ interface DescribeFileErrorResponse {
   error: string;
 }
 
-/**
- * Extract text content from Claude SDK response messages
- */
-async function extractTextFromStream(
-  // eslint-disable-next-line @typescript-eslint/no-explicit-any
-  stream: AsyncIterable<any>
-): Promise<string> {
-  let responseText = '';
-
-  for await (const msg of stream) {
-    if (msg.type === 'assistant' && msg.message?.content) {
-      const blocks = msg.message.content as Array<{ type: string; text?: string }>;
-      for (const block of blocks) {
-        if (block.type === 'text' && block.text) {
-          responseText += block.text;
-        }
-      }
-    } else if (msg.type === 'result' && msg.subtype === 'success') {
-      responseText = msg.result || responseText;
-    }
-  }
-
-  return responseText;
-}
-
 /**
  * Create the describe-file request handler
  *
@@ -94,7 +70,7 @@ export function createDescribeFileHandler(
         return;
       }
 
-      logger.info(`[DescribeFile] Starting description generation for: ${filePath}`);
+      logger.info(`Starting description generation for: ${filePath}`);
 
       // Resolve the path for logging and cwd derivation
       const resolvedPath = secureFs.resolvePath(filePath);
@@ -109,7 +85,7 @@ export function createDescribeFileHandler(
       } catch (readError) {
         // Path not allowed - return 403 Forbidden
         if (readError instanceof PathNotAllowedError) {
-          logger.warn(`[DescribeFile] Path not allowed: ${filePath}`);
+          logger.warn(`Path not allowed: ${filePath}`);
           const response: DescribeFileErrorResponse = {
             success: false,
             error: 'File path is not within the allowed directory',
@@ -125,7 +101,7 @@ export function createDescribeFileHandler(
           'code' in readError &&
           readError.code === 'ENOENT'
         ) {
-          logger.warn(`[DescribeFile] File not found: ${resolvedPath}`);
+          logger.warn(`File not found: ${resolvedPath}`);
           const response: DescribeFileErrorResponse = {
             success: false,
             error: `File not found: ${filePath}`,
@@ -135,7 +111,7 @@ export function createDescribeFileHandler(
         }
 
         const errorMessage = readError instanceof Error ? readError.message : 'Unknown error';
-        logger.error(`[DescribeFile] Failed to read file: ${errorMessage}`);
+        logger.error(`Failed to read file: ${errorMessage}`);
         const response: DescribeFileErrorResponse = {
           success: false,
           error: `Failed to read file: ${errorMessage}`,
@@ -156,16 +132,14 @@ export function createDescribeFileHandler(
 
       // Build prompt with file content passed as structured data
       // The file content is included directly, not via tool invocation
-      const instructionText = `Analyze the following file and provide a 1-2 sentence description suitable for use as context in an AI coding assistant. Focus on what the file contains, its purpose, and why an AI agent might want to use this context in the future (e.g., "API documentation for the authentication endpoints", "Configuration file for database connections", "Coding style guidelines for the project").
+      const prompt = `Analyze the following file and provide a 1-2 sentence description suitable for use as context in an AI coding assistant. Focus on what the file contains, its purpose, and why an AI agent might want to use this context in the future (e.g., "API documentation for the authentication endpoints", "Configuration file for database connections", "Coding style guidelines for the project").
 
 Respond with ONLY the description text, no additional formatting, preamble, or explanation.
 
-File: ${fileName}${truncated ? ' (truncated)' : ''}`;
+File: ${fileName}${truncated ? ' (truncated)' : ''}
 
-      const promptContent = [
-        { type: 'text' as const, text: instructionText },
-        { type: 'text' as const, text: `\n\n--- FILE CONTENT ---\n${contentToAnalyze}` },
-      ];
+--- FILE CONTENT ---
+${contentToAnalyze}`;
 
       // Use the file's directory as the working directory
       const cwd = path.dirname(resolvedPath);
@@ -177,30 +151,29 @@ File: ${fileName}${truncated ? ' (truncated)' : ''}`;
         '[DescribeFile]'
       );
 
-      // Use centralized SDK options with proper cwd validation
-      // No tools needed since we're passing file content directly
-      const sdkOptions = createCustomOptions({
+      // Get model from phase settings
+      const settings = await settingsService?.getGlobalSettings();
+      logger.info(`Raw phaseModels from settings:`, JSON.stringify(settings?.phaseModels, null, 2));
+      const phaseModelEntry =
+        settings?.phaseModels?.fileDescriptionModel || DEFAULT_PHASE_MODELS.fileDescriptionModel;
+      logger.info(`fileDescriptionModel entry:`, JSON.stringify(phaseModelEntry));
+      const { model, thinkingLevel } = resolvePhaseModel(phaseModelEntry);
+
+      logger.info(`Resolved model: ${model}, thinkingLevel: ${thinkingLevel}`);
+
+      // Use simpleQuery - provider abstraction handles routing to correct provider
+      const result = await simpleQuery({
+        prompt,
+        model,
         cwd,
-        model: CLAUDE_MODEL_MAP.haiku,
         maxTurns: 1,
         allowedTools: [],
-        autoLoadClaudeMd,
-        sandbox: { enabled: true, autoAllowBashIfSandboxed: true },
+        thinkingLevel,
+        readOnly: true, // File description only reads, doesn't write
+        settingSources: autoLoadClaudeMd ? ['user', 'project', 'local'] : undefined,
       });
 
-      const promptGenerator = (async function* () {
-        yield {
-          type: 'user' as const,
-          session_id: '',
-          message: { role: 'user' as const, content: promptContent },
-          parent_tool_use_id: null,
-        };
-      })();
-
-      const stream = query({ prompt: promptGenerator, options: sdkOptions });
-
-      // Extract the description from the response
-      const description = await extractTextFromStream(stream);
+      const description = result.text;
 
       if (!description || description.trim().length === 0) {
         logger.warn('Received empty response from Claude');

apps/server/src/routes/context/routes/describe-image.ts

@@ -1,8 +1,9 @@
 /**
  * POST /context/describe-image endpoint - Generate description for an image
  *
- * Uses Claude Haiku to analyze an image and generate a concise description
- * suitable for context file metadata.
+ * Uses AI to analyze an image and generate a concise description
+ * suitable for context file metadata. Model is configurable via
+ * phaseModels.imageDescriptionModel in settings (defaults to Haiku).
  *
  * IMPORTANT:
  * The agent runner (chat/auto-mode) sends images as multi-part content blocks (base64 image blocks),
@@ -11,10 +12,10 @@
  */
 
 import type { Request, Response } from 'express';
-import { query } from '@anthropic-ai/claude-agent-sdk';
 import { createLogger, readImageAsBase64 } from '@automaker/utils';
-import { CLAUDE_MODEL_MAP } from '@automaker/types';
-import { createCustomOptions } from '../../../lib/sdk-options.js';
+import { DEFAULT_PHASE_MODELS, isCursorModel } from '@automaker/types';
+import { resolvePhaseModel } from '@automaker/model-resolver';
+import { simpleQuery } from '../../../providers/simple-query-service.js';
 import * as secureFs from '../../../lib/secure-fs.js';
 import * as path from 'path';
 import type { SettingsService } from '../../../services/settings-service.js';
@@ -175,57 +176,10 @@ function mapDescribeImageError(rawMessage: string | undefined): {
   return baseResponse;
 }
 
-/**
- * Extract text content from Claude SDK response messages and log high-signal stream events.
- */
-async function extractTextFromStream(
-  // eslint-disable-next-line @typescript-eslint/no-explicit-any
-  stream: AsyncIterable<any>,
-  requestId: string
-): Promise<string> {
-  let responseText = '';
-  let messageCount = 0;
-
-  logger.info(`[${requestId}] [Stream] Begin reading SDK stream...`);
-
-  for await (const msg of stream) {
-    messageCount++;
-    const msgType = msg?.type;
-    const msgSubtype = msg?.subtype;
-
-    // Keep this concise but informative. Full error object is logged in catch blocks.
-    logger.info(
-      `[${requestId}] [Stream] #${messageCount} type=${String(msgType)} subtype=${String(msgSubtype ?? '')}`
-    );
-
-    if (msgType === 'assistant' && msg.message?.content) {
-      const blocks = msg.message.content as Array<{ type: string; text?: string }>;
-      logger.info(`[${requestId}] [Stream] assistant blocks=${blocks.length}`);
-      for (const block of blocks) {
-        if (block.type === 'text' && block.text) {
-          responseText += block.text;
-        }
-      }
-    }
-
-    if (msgType === 'result' && msgSubtype === 'success') {
-      if (typeof msg.result === 'string' && msg.result.length > 0) {
-        responseText = msg.result;
-      }
-    }
-  }
-
-  logger.info(
-    `[${requestId}] [Stream] End of stream. messages=${messageCount} textLength=${responseText.length}`
-  );
-
-  return responseText;
-}
-
 /**
  * Create the describe-image request handler
  *
- * Uses Claude SDK query with multi-part content blocks to include the image (base64),
+ * Uses the provider abstraction with multi-part content blocks to include the image (base64),
  * matching the agent runner behavior.
  *
  * @param settingsService - Optional settings service for loading autoLoadClaudeMd setting
@@ -306,27 +260,6 @@ export function createDescribeImageHandler(
         `[${requestId}] image meta filename=${imageData.filename} mime=${imageData.mimeType} base64Len=${base64Length} estBytes=${estimatedBytes}`
       );
 
-      // Build multi-part prompt with image block (no Read tool required)
-      const instructionText =
-        `Describe this image in 1-2 sentences suitable for use as context in an AI coding assistant. ` +
-        `Focus on what the image shows and its purpose (e.g., "UI mockup showing login form with email/password fields", ` +
-        `"Architecture diagram of microservices", "Screenshot of error message in terminal").\n\n` +
-        `Respond with ONLY the description text, no additional formatting, preamble, or explanation.`;
-
-      const promptContent = [
-        { type: 'text' as const, text: instructionText },
-        {
-          type: 'image' as const,
-          source: {
-            type: 'base64' as const,
-            media_type: imageData.mimeType,
-            data: imageData.base64,
-          },
-        },
-      ];
-
-      logger.info(`[${requestId}] Built multi-part prompt blocks=${promptContent.length}`);
-
       const cwd = path.dirname(actualPath);
       logger.info(`[${requestId}] Using cwd=${cwd}`);
 
@@ -337,43 +270,67 @@ export function createDescribeImageHandler(
         '[DescribeImage]'
       );
 
-      // Use the same centralized option builder used across the server (validates cwd)
-      const sdkOptions = createCustomOptions({
+      // Get model from phase settings
+      const settings = await settingsService?.getGlobalSettings();
+      const phaseModelEntry =
+        settings?.phaseModels?.imageDescriptionModel || DEFAULT_PHASE_MODELS.imageDescriptionModel;
+      const { model, thinkingLevel } = resolvePhaseModel(phaseModelEntry);
+
+      logger.info(`[${requestId}] Using model: ${model}`);
+
+      // Build the instruction text
+      const instructionText =
+        `Describe this image in 1-2 sentences suitable for use as context in an AI coding assistant. ` +
+        `Focus on what the image shows and its purpose (e.g., "UI mockup showing login form with email/password fields", ` +
+        `"Architecture diagram of microservices", "Screenshot of error message in terminal").\n\n` +
+        `Respond with ONLY the description text, no additional formatting, preamble, or explanation.`;
+
+      // Build prompt based on provider capability
+      // Some providers (like Cursor) may not support image content blocks
+      let prompt: string | Array<{ type: string; text?: string; source?: object }>;
+
+      if (isCursorModel(model)) {
+        // Cursor may not support base64 image blocks directly
+        // Use text prompt with image path reference
+        logger.info(`[${requestId}] Using text prompt for Cursor model`);
+        prompt = `${instructionText}\n\nImage file: ${actualPath}\nMIME type: ${imageData.mimeType}`;
+      } else {
+        // Claude and other vision-capable models support multi-part prompts with images
+        logger.info(`[${requestId}] Using multi-part prompt with image block`);
+        prompt = [
+          { type: 'text', text: instructionText },
+          {
+            type: 'image',
+            source: {
+              type: 'base64',
+              media_type: imageData.mimeType,
+              data: imageData.base64,
+            },
+          },
+        ];
+      }
+
+      logger.info(`[${requestId}] Calling simpleQuery...`);
+      const queryStart = Date.now();
+
+      // Use simpleQuery - provider abstraction handles routing
+      const result = await simpleQuery({
+        prompt,
+        model,
         cwd,
-        model: CLAUDE_MODEL_MAP.haiku,
         maxTurns: 1,
-        allowedTools: [],
-        autoLoadClaudeMd,
-        sandbox: { enabled: true, autoAllowBashIfSandboxed: true },
+        allowedTools: isCursorModel(model) ? ['Read'] : [], // Allow Read for Cursor to read image if needed
+        thinkingLevel,
+        readOnly: true, // Image description only reads, doesn't write
+        settingSources: autoLoadClaudeMd ? ['user', 'project', 'local'] : undefined,
       });
 
-      logger.info(
-        `[${requestId}] SDK options model=${sdkOptions.model} maxTurns=${sdkOptions.maxTurns} allowedTools=${JSON.stringify(
-          sdkOptions.allowedTools
-        )} sandbox=${JSON.stringify(sdkOptions.sandbox)}`
-      );
+      logger.info(`[${requestId}] simpleQuery completed in ${Date.now() - queryStart}ms`);
 
-      const promptGenerator = (async function* () {
-        yield {
-          type: 'user' as const,
-          session_id: '',
-          message: { role: 'user' as const, content: promptContent },
-          parent_tool_use_id: null,
-        };
-      })();
-
-      logger.info(`[${requestId}] Calling query()...`);
-      const queryStart = Date.now();
-      const stream = query({ prompt: promptGenerator, options: sdkOptions });
-      logger.info(`[${requestId}] query() returned stream in ${Date.now() - queryStart}ms`);
-
-      // Extract the description from the response
-      const extractStart = Date.now();
-      const description = await extractTextFromStream(stream, requestId);
-      logger.info(`[${requestId}] extractMs=${Date.now() - extractStart}`);
+      const description = result.text;
 
       if (!description || description.trim().length === 0) {
-        logger.warn(`[${requestId}] Received empty response from Claude`);
+        logger.warn(`[${requestId}] Received empty response from AI`);
         const response: DescribeImageErrorResponse = {
           success: false,
           error: 'Failed to generate description - empty response',

apps/server/src/routes/enhance-prompt/routes/enhance.ts

@@ -1,15 +1,16 @@
 /**
  * POST /enhance-prompt endpoint - Enhance user input text
  *
- * Uses Claude AI to enhance text based on the specified enhancement mode.
- * Supports modes: improve, technical, simplify, acceptance
+ * Uses the provider abstraction to enhance text based on the specified
+ * enhancement mode. Works with any configured provider (Claude, Cursor, etc.).
+ * Supports modes: improve, technical, simplify, acceptance, ux-reviewer
  */
 
 import type { Request, Response } from 'express';
-import { query } from '@anthropic-ai/claude-agent-sdk';
 import { createLogger } from '@automaker/utils';
 import { resolveModelString } from '@automaker/model-resolver';
-import { CLAUDE_MODEL_MAP } from '@automaker/types';
+import { CLAUDE_MODEL_MAP, type ThinkingLevel } from '@automaker/types';
+import { simpleQuery } from '../../../providers/simple-query-service.js';
 import type { SettingsService } from '../../../services/settings-service.js';
 import { getPromptCustomization } from '../../../lib/settings-helpers.js';
 import {
@@ -30,6 +31,8 @@ interface EnhanceRequestBody {
   enhancementMode: string;
   /** Optional model override */
   model?: string;
+  /** Optional thinking level for Claude models */
+  thinkingLevel?: ThinkingLevel;
 }
 
 /**
@@ -48,39 +51,6 @@ interface EnhanceErrorResponse {
   error: string;
 }
 
-/**
- * Extract text content from Claude SDK response messages
- *
- * @param stream - The async iterable from the query function
- * @returns The extracted text content
- */
-async function extractTextFromStream(
-  stream: AsyncIterable<{
-    type: string;
-    subtype?: string;
-    result?: string;
-    message?: {
-      content?: Array<{ type: string; text?: string }>;
-    };
-  }>
-): Promise<string> {
-  let responseText = '';
-
-  for await (const msg of stream) {
-    if (msg.type === 'assistant' && msg.message?.content) {
-      for (const block of msg.message.content) {
-        if (block.type === 'text' && block.text) {
-          responseText += block.text;
-        }
-      }
-    } else if (msg.type === 'result' && msg.subtype === 'success') {
-      responseText = msg.result || responseText;
-    }
-  }
-
-  return responseText;
-}
-
 /**
  * Create the enhance request handler
  *
@@ -92,7 +62,8 @@ export function createEnhanceHandler(
 ): (req: Request, res: Response) => Promise<void> {
   return async (req: Request, res: Response): Promise<void> => {
     try {
-      const { originalText, enhancementMode, model } = req.body as EnhanceRequestBody;
+      const { originalText, enhancementMode, model, thinkingLevel } =
+        req.body as EnhanceRequestBody;
 
       // Validate required fields
       if (!originalText || typeof originalText !== 'string') {
@@ -141,13 +112,13 @@ export function createEnhanceHandler(
         technical: prompts.enhancement.technicalSystemPrompt,
         simplify: prompts.enhancement.simplifySystemPrompt,
         acceptance: prompts.enhancement.acceptanceSystemPrompt,
+        'ux-reviewer': prompts.enhancement.uxReviewerSystemPrompt,
       };
       const systemPrompt = systemPromptMap[validMode];
 
       logger.debug(`Using ${validMode} system prompt (length: ${systemPrompt.length} chars)`);
 
       // Build the user prompt with few-shot examples
-      // This helps the model understand this is text transformation, not a coding task
       const userPrompt = buildUserPrompt(validMode, trimmedText, true);
 
       // Resolve the model - use the passed model, default to sonnet for quality
@@ -155,24 +126,23 @@ export function createEnhanceHandler(
 
       logger.debug(`Using model: ${resolvedModel}`);
 
-      // Call Claude SDK with minimal configuration for text transformation
-      // Key: no tools, just text completion
-      const stream = query({
-        prompt: userPrompt,
-        options: {
-          model: resolvedModel,
-          systemPrompt,
-          maxTurns: 1,
-          allowedTools: [],
-          permissionMode: 'acceptEdits',
-        },
+      // Use simpleQuery - provider abstraction handles routing to correct provider
+      // The system prompt is combined with user prompt since some providers
+      // don't have a separate system prompt concept
+      const result = await simpleQuery({
+        prompt: `${systemPrompt}\n\n${userPrompt}`,
+        model: resolvedModel,
+        cwd: process.cwd(), // Enhancement doesn't need a specific working directory
+        maxTurns: 1,
+        allowedTools: [],
+        thinkingLevel,
+        readOnly: true, // Prompt enhancement only generates text, doesn't write files
       });
 
-      // Extract the enhanced text from the response
-      const enhancedText = await extractTextFromStream(stream);
+      const enhancedText = result.text;
 
       if (!enhancedText || enhancedText.trim().length === 0) {
-        logger.warn('Received empty response from Claude');
+        logger.warn('Received empty response from AI');
         const response: EnhanceErrorResponse = {
           success: false,
           error: 'Failed to generate enhanced text - empty response',

apps/server/src/routes/features/index.ts

@@ -9,8 +9,10 @@ import { createListHandler } from './routes/list.js';
 import { createGetHandler } from './routes/get.js';
 import { createCreateHandler } from './routes/create.js';
 import { createUpdateHandler } from './routes/update.js';
+import { createBulkUpdateHandler } from './routes/bulk-update.js';
+import { createBulkDeleteHandler } from './routes/bulk-delete.js';
 import { createDeleteHandler } from './routes/delete.js';
-import { createAgentOutputHandler } from './routes/agent-output.js';
+import { createAgentOutputHandler, createRawOutputHandler } from './routes/agent-output.js';
 import { createGenerateTitleHandler } from './routes/generate-title.js';
 
 export function createFeaturesRoutes(featureLoader: FeatureLoader): Router {
@@ -20,8 +22,19 @@ export function createFeaturesRoutes(featureLoader: FeatureLoader): Router {
   router.post('/get', validatePathParams('projectPath'), createGetHandler(featureLoader));
   router.post('/create', validatePathParams('projectPath'), createCreateHandler(featureLoader));
   router.post('/update', validatePathParams('projectPath'), createUpdateHandler(featureLoader));
+  router.post(
+    '/bulk-update',
+    validatePathParams('projectPath'),
+    createBulkUpdateHandler(featureLoader)
+  );
+  router.post(
+    '/bulk-delete',
+    validatePathParams('projectPath'),
+    createBulkDeleteHandler(featureLoader)
+  );
   router.post('/delete', validatePathParams('projectPath'), createDeleteHandler(featureLoader));
   router.post('/agent-output', createAgentOutputHandler(featureLoader));
+  router.post('/raw-output', createRawOutputHandler(featureLoader));
   router.post('/generate-title', createGenerateTitleHandler());
 
   return router;

apps/server/src/routes/features/routes/agent-output.ts

@@ -1,5 +1,6 @@
 /**
  * POST /agent-output endpoint - Get agent output for a feature
+ * POST /raw-output endpoint - Get raw JSONL output for debugging
  */
 
 import type { Request, Response } from 'express';
@@ -30,3 +31,31 @@ export function createAgentOutputHandler(featureLoader: FeatureLoader) {
     }
   };
 }
+
+/**
+ * Handler for getting raw JSONL output for debugging
+ */
+export function createRawOutputHandler(featureLoader: FeatureLoader) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, featureId } = req.body as {
+        projectPath: string;
+        featureId: string;
+      };
+
+      if (!projectPath || !featureId) {
+        res.status(400).json({
+          success: false,
+          error: 'projectPath and featureId are required',
+        });
+        return;
+      }
+
+      const content = await featureLoader.getRawOutput(projectPath, featureId);
+      res.json({ success: true, content });
+    } catch (error) {
+      logError(error, 'Get raw output failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/features/routes/bulk-delete.ts

@@ -0,0 +1,61 @@
+/**
+ * POST /bulk-delete endpoint - Delete multiple features at once
+ */
+
+import type { Request, Response } from 'express';
+import { FeatureLoader } from '../../../services/feature-loader.js';
+import { getErrorMessage, logError } from '../common.js';
+
+interface BulkDeleteRequest {
+  projectPath: string;
+  featureIds: string[];
+}
+
+interface BulkDeleteResult {
+  featureId: string;
+  success: boolean;
+  error?: string;
+}
+
+export function createBulkDeleteHandler(featureLoader: FeatureLoader) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, featureIds } = req.body as BulkDeleteRequest;
+
+      if (!projectPath || !featureIds || !Array.isArray(featureIds) || featureIds.length === 0) {
+        res.status(400).json({
+          success: false,
+          error: 'projectPath and featureIds (non-empty array) are required',
+        });
+        return;
+      }
+
+      const results = await Promise.all(
+        featureIds.map(async (featureId) => {
+          const success = await featureLoader.delete(projectPath, featureId);
+          if (success) {
+            return { featureId, success: true };
+          }
+          return {
+            featureId,
+            success: false,
+            error: 'Deletion failed. Check server logs for details.',
+          };
+        })
+      );
+
+      const successCount = results.reduce((count, r) => count + (r.success ? 1 : 0), 0);
+      const failureCount = results.length - successCount;
+
+      res.json({
+        success: failureCount === 0,
+        deletedCount: successCount,
+        failedCount: failureCount,
+        results,
+      });
+    } catch (error) {
+      logError(error, 'Bulk delete features failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/features/routes/bulk-update.ts

@@ -0,0 +1,75 @@
+/**
+ * POST /bulk-update endpoint - Update multiple features at once
+ */
+
+import type { Request, Response } from 'express';
+import { FeatureLoader } from '../../../services/feature-loader.js';
+import type { Feature } from '@automaker/types';
+import { getErrorMessage, logError } from '../common.js';
+
+interface BulkUpdateRequest {
+  projectPath: string;
+  featureIds: string[];
+  updates: Partial<Feature>;
+}
+
+interface BulkUpdateResult {
+  featureId: string;
+  success: boolean;
+  error?: string;
+}
+
+export function createBulkUpdateHandler(featureLoader: FeatureLoader) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, featureIds, updates } = req.body as BulkUpdateRequest;
+
+      if (!projectPath || !featureIds || !Array.isArray(featureIds) || featureIds.length === 0) {
+        res.status(400).json({
+          success: false,
+          error: 'projectPath and featureIds (non-empty array) are required',
+        });
+        return;
+      }
+
+      if (!updates || Object.keys(updates).length === 0) {
+        res.status(400).json({
+          success: false,
+          error: 'updates object with at least one field is required',
+        });
+        return;
+      }
+
+      const results: BulkUpdateResult[] = [];
+      const updatedFeatures: Feature[] = [];
+
+      for (const featureId of featureIds) {
+        try {
+          const updated = await featureLoader.update(projectPath, featureId, updates);
+          results.push({ featureId, success: true });
+          updatedFeatures.push(updated);
+        } catch (error) {
+          results.push({
+            featureId,
+            success: false,
+            error: getErrorMessage(error),
+          });
+        }
+      }
+
+      const successCount = results.filter((r) => r.success).length;
+      const failureCount = results.filter((r) => !r.success).length;
+
+      res.json({
+        success: failureCount === 0,
+        updatedCount: successCount,
+        failedCount: failureCount,
+        results,
+        features: updatedFeatures,
+      });
+    } catch (error) {
+      logError(error, 'Bulk update features failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/features/routes/generate-title.ts

@@ -1,13 +1,14 @@
 /**
  * POST /features/generate-title endpoint - Generate a concise title from description
  *
- * Uses Claude Haiku to generate a short, descriptive title from feature description.
+ * Uses the provider abstraction to generate a short, descriptive title
+ * from a feature description. Works with any configured provider (Claude, Cursor, etc.).
  */
 
 import type { Request, Response } from 'express';
-import { query } from '@anthropic-ai/claude-agent-sdk';
 import { createLogger } from '@automaker/utils';
 import { CLAUDE_MODEL_MAP } from '@automaker/model-resolver';
+import { simpleQuery } from '../../../providers/simple-query-service.js';
 
 const logger = createLogger('GenerateTitle');
 
@@ -34,33 +35,6 @@ Rules:
 - No quotes, periods, or extra formatting
 - Capture the essence of the feature in a scannable way`;
 
-async function extractTextFromStream(
-  stream: AsyncIterable<{
-    type: string;
-    subtype?: string;
-    result?: string;
-    message?: {
-      content?: Array<{ type: string; text?: string }>;
-    };
-  }>
-): Promise<string> {
-  let responseText = '';
-
-  for await (const msg of stream) {
-    if (msg.type === 'assistant' && msg.message?.content) {
-      for (const block of msg.message.content) {
-        if (block.type === 'text' && block.text) {
-          responseText += block.text;
-        }
-      }
-    } else if (msg.type === 'result' && msg.subtype === 'success') {
-      responseText = msg.result || responseText;
-    }
-  }
-
-  return responseText;
-}
-
 export function createGenerateTitleHandler(): (req: Request, res: Response) => Promise<void> {
   return async (req: Request, res: Response): Promise<void> => {
     try {
@@ -89,21 +63,19 @@ export function createGenerateTitleHandler(): (req: Request, res: Response) => P
 
       const userPrompt = `Generate a concise title for this feature:\n\n${trimmedDescription}`;
 
-      const stream = query({
-        prompt: userPrompt,
-        options: {
-          model: CLAUDE_MODEL_MAP.haiku,
-          systemPrompt: SYSTEM_PROMPT,
-          maxTurns: 1,
-          allowedTools: [],
-          permissionMode: 'acceptEdits',
-        },
+      // Use simpleQuery - provider abstraction handles all the streaming/extraction
+      const result = await simpleQuery({
+        prompt: `${SYSTEM_PROMPT}\n\n${userPrompt}`,
+        model: CLAUDE_MODEL_MAP.haiku,
+        cwd: process.cwd(),
+        maxTurns: 1,
+        allowedTools: [],
       });
 
-      const title = await extractTextFromStream(stream);
+      const title = result.text;
 
       if (!title || title.trim().length === 0) {
-        logger.warn('Received empty response from Claude');
+        logger.warn('Received empty response from AI');
         const response: GenerateTitleErrorResponse = {
           success: false,
           error: 'Failed to generate title - empty response',

apps/server/src/routes/features/routes/update.ts

@@ -10,10 +10,20 @@ import { getErrorMessage, logError } from '../common.js';
 export function createUpdateHandler(featureLoader: FeatureLoader) {
   return async (req: Request, res: Response): Promise<void> => {
     try {
-      const { projectPath, featureId, updates } = req.body as {
+      const {
+        projectPath,
+        featureId,
+        updates,
+        descriptionHistorySource,
+        enhancementMode,
+        preEnhancementDescription,
+      } = req.body as {
         projectPath: string;
         featureId: string;
         updates: Partial<Feature>;
+        descriptionHistorySource?: 'enhance' | 'edit';
+        enhancementMode?: 'improve' | 'technical' | 'simplify' | 'acceptance' | 'ux-reviewer';
+        preEnhancementDescription?: string;
       };
 
       if (!projectPath || !featureId || !updates) {
@@ -24,7 +34,14 @@ export function createUpdateHandler(featureLoader: FeatureLoader) {
         return;
       }
 
-      const updated = await featureLoader.update(projectPath, featureId, updates);
+      const updated = await featureLoader.update(
+        projectPath,
+        featureId,
+        updates,
+        descriptionHistorySource,
+        enhancementMode,
+        preEnhancementDescription
+      );
       res.json({ success: true, feature: updated });
     } catch (error) {
       logError(error, 'Update feature failed');

apps/server/src/routes/github/routes/common.ts

@@ -4,6 +4,9 @@
 
 import { exec } from 'child_process';
 import { promisify } from 'util';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('GitHub');
 
 export const execAsync = promisify(exec);
 
@@ -31,5 +34,5 @@ export function getErrorMessage(error: unknown): string {
 }
 
 export function logError(error: unknown, context: string): void {
-  console.error(`[GitHub] ${context}:`, error);
+  logger.error(`${context}:`, error);
 }

apps/server/src/routes/github/routes/list-issues.ts

@@ -6,6 +6,9 @@ import { spawn } from 'child_process';
 import type { Request, Response } from 'express';
 import { execAsync, execEnv, getErrorMessage, logError } from './common.js';
 import { checkGitHubRemote } from './check-github-remote.js';
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('ListIssues');
 
 export interface GitHubLabel {
   name: string;
@@ -179,7 +182,7 @@ async function fetchLinkedPRs(
       }
     } catch (error) {
       // If GraphQL fails, continue without linked PRs
-      console.warn(
+      logger.warn(
         'Failed to fetch linked PRs via GraphQL:',
         error instanceof Error ? error.message : error
       );

apps/server/src/routes/github/routes/validate-issue.ts

@@ -1,22 +1,27 @@
 /**
- * POST /validate-issue endpoint - Validate a GitHub issue using Claude SDK (async)
+ * POST /validate-issue endpoint - Validate a GitHub issue using provider abstraction (async)
  *
  * Scans the codebase to determine if an issue is valid, invalid, or needs clarification.
  * Runs asynchronously and emits events for progress and completion.
+ * Supports both Claude models and Cursor models.
  */
 
 import type { Request, Response } from 'express';
-import { query } from '@anthropic-ai/claude-agent-sdk';
 import type { EventEmitter } from '../../../lib/events.js';
 import type {
   IssueValidationResult,
   IssueValidationEvent,
-  AgentModel,
+  ModelAlias,
+  CursorModelId,
   GitHubComment,
   LinkedPRInfo,
+  ThinkingLevel,
 } from '@automaker/types';
-import { createSuggestionsOptions } from '../../../lib/sdk-options.js';
+import { isCursorModel, DEFAULT_PHASE_MODELS } from '@automaker/types';
+import { resolvePhaseModel } from '@automaker/model-resolver';
+import { extractJson } from '../../../lib/json-extractor.js';
 import { writeValidation } from '../../../lib/validation-storage.js';
+import { streamingQuery } from '../../../providers/simple-query-service.js';
 import {
   issueValidationSchema,
   ISSUE_VALIDATION_SYSTEM_PROMPT,
@@ -34,8 +39,8 @@ import {
 import type { SettingsService } from '../../../services/settings-service.js';
 import { getAutoLoadClaudeMdSetting } from '../../../lib/settings-helpers.js';
 
-/** Valid model values for validation */
-const VALID_MODELS: readonly AgentModel[] = ['opus', 'sonnet', 'haiku'] as const;
+/** Valid Claude model values for validation */
+const VALID_CLAUDE_MODELS: readonly ModelAlias[] = ['opus', 'sonnet', 'haiku'] as const;
 
 /**
  * Request body for issue validation
@@ -46,8 +51,10 @@ interface ValidateIssueRequestBody {
   issueTitle: string;
   issueBody: string;
   issueLabels?: string[];
-  /** Model to use for validation (opus, sonnet, haiku) */
-  model?: AgentModel;
+  /** Model to use for validation (opus, sonnet, haiku, or cursor model IDs) */
+  model?: ModelAlias | CursorModelId;
+  /** Thinking level for Claude models (ignored for Cursor models) */
+  thinkingLevel?: ThinkingLevel;
   /** Comments to include in validation analysis */
   comments?: GitHubComment[];
   /** Linked pull requests for this issue */
@@ -59,6 +66,7 @@ interface ValidateIssueRequestBody {
  *
  * Emits events for start, progress, complete, and error.
  * Stores result on completion.
+ * Supports both Claude models (with structured output) and Cursor models (with JSON parsing).
  */
 async function runValidation(
   projectPath: string,
@@ -66,12 +74,13 @@ async function runValidation(
   issueTitle: string,
   issueBody: string,
   issueLabels: string[] | undefined,
-  model: AgentModel,
+  model: ModelAlias | CursorModelId,
   events: EventEmitter,
   abortController: AbortController,
   settingsService?: SettingsService,
   comments?: ValidationComment[],
-  linkedPRs?: ValidationLinkedPR[]
+  linkedPRs?: ValidationLinkedPR[],
+  thinkingLevel?: ThinkingLevel
 ): Promise<void> {
   // Emit start event
   const startEvent: IssueValidationEvent = {
@@ -91,7 +100,7 @@ async function runValidation(
 
   try {
     // Build the prompt (include comments and linked PRs if provided)
-    const prompt = buildValidationPrompt(
+    const basePrompt = buildValidationPrompt(
       issueNumber,
       issueTitle,
       issueBody,
@@ -100,6 +109,28 @@ async function runValidation(
       linkedPRs
     );
 
+    let responseText = '';
+
+    // Determine if we should use structured output (Claude supports it, Cursor doesn't)
+    const useStructuredOutput = !isCursorModel(model);
+
+    // Build the final prompt - for Cursor, include system prompt and JSON schema instructions
+    let finalPrompt = basePrompt;
+    if (!useStructuredOutput) {
+      finalPrompt = `${ISSUE_VALIDATION_SYSTEM_PROMPT}
+
+CRITICAL INSTRUCTIONS:
+1. DO NOT write any files. Return the JSON in your response only.
+2. Respond with ONLY a JSON object - no explanations, no markdown, just raw JSON.
+3. The JSON must match this exact schema:
+
+${JSON.stringify(issueValidationSchema, null, 2)}
+
+Your entire response should be valid JSON starting with { and ending with }. No text before or after.
+
+${basePrompt}`;
+    }
+
     // Load autoLoadClaudeMd setting
     const autoLoadClaudeMd = await getAutoLoadClaudeMdSetting(
       projectPath,
@@ -107,64 +138,65 @@ async function runValidation(
       '[ValidateIssue]'
     );
 
-    // Create SDK options with structured output and abort controller
-    const options = createSuggestionsOptions({
+    // Use thinkingLevel from request if provided, otherwise fall back to settings
+    let effectiveThinkingLevel: ThinkingLevel | undefined = thinkingLevel;
+    if (!effectiveThinkingLevel) {
+      const settings = await settingsService?.getGlobalSettings();
+      const phaseModelEntry =
+        settings?.phaseModels?.validationModel || DEFAULT_PHASE_MODELS.validationModel;
+      const resolved = resolvePhaseModel(phaseModelEntry);
+      effectiveThinkingLevel = resolved.thinkingLevel;
+    }
+
+    logger.info(`Using model: ${model}`);
+
+    // Use streamingQuery with event callbacks
+    const result = await streamingQuery({
+      prompt: finalPrompt,
+      model: model as string,
       cwd: projectPath,
-      model,
-      systemPrompt: ISSUE_VALIDATION_SYSTEM_PROMPT,
+      systemPrompt: useStructuredOutput ? ISSUE_VALIDATION_SYSTEM_PROMPT : undefined,
       abortController,
-      autoLoadClaudeMd,
-      outputFormat: {
-        type: 'json_schema',
-        schema: issueValidationSchema as Record<string, unknown>,
+      thinkingLevel: effectiveThinkingLevel,
+      readOnly: true, // Issue validation only reads code, doesn't write
+      settingSources: autoLoadClaudeMd ? ['user', 'project', 'local'] : undefined,
+      outputFormat: useStructuredOutput
+        ? {
+            type: 'json_schema',
+            schema: issueValidationSchema as Record<string, unknown>,
+          }
+        : undefined,
+      onText: (text) => {
+        responseText += text;
+        // Emit progress event
+        const progressEvent: IssueValidationEvent = {
+          type: 'issue_validation_progress',
+          issueNumber,
+          content: text,
+          projectPath,
+        };
+        events.emit('issue-validation:event', progressEvent);
       },
     });
 
-    // Execute the query
-    const stream = query({ prompt, options });
-    let validationResult: IssueValidationResult | null = null;
-
-    for await (const msg of stream) {
-      // Emit progress events for assistant text
-      if (msg.type === 'assistant' && msg.message?.content) {
-        for (const block of msg.message.content) {
-          if (block.type === 'text') {
-            const progressEvent: IssueValidationEvent = {
-              type: 'issue_validation_progress',
-              issueNumber,
-              content: block.text,
-              projectPath,
-            };
-            events.emit('issue-validation:event', progressEvent);
-          }
-        }
-      }
-
-      // Extract structured output on success
-      if (msg.type === 'result' && msg.subtype === 'success') {
-        const resultMsg = msg as { structured_output?: IssueValidationResult };
-        if (resultMsg.structured_output) {
-          validationResult = resultMsg.structured_output;
-        }
-      }
-
-      // Handle errors
-      if (msg.type === 'result') {
-        const resultMsg = msg as { subtype?: string };
-        if (resultMsg.subtype === 'error_max_structured_output_retries') {
-          logger.error('Failed to produce valid structured output after retries');
-          throw new Error('Could not produce valid validation output');
-        }
-      }
-    }
-
     // Clear timeout
     clearTimeout(timeoutId);
 
-    // Require structured output
+    // Get validation result from structured output or parse from text
+    let validationResult: IssueValidationResult | null = null;
+
+    if (result.structured_output) {
+      validationResult = result.structured_output as unknown as IssueValidationResult;
+      logger.debug('Received structured output:', validationResult);
+    } else if (responseText) {
+      // Parse JSON from response text
+      validationResult = extractJson<IssueValidationResult>(responseText, { logger });
+    }
+
+    // Require validation result
     if (!validationResult) {
-      logger.error('No structured output received from Claude SDK');
-      throw new Error('Validation failed: no structured output received');
+      logger.error('No validation result received from AI provider');
+      throw new Error('Validation failed: no valid result received');
     }
 
     logger.info(`Issue #${issueNumber} validation complete: ${validationResult.verdict}`);
@@ -210,7 +242,7 @@ async function runValidation(
 /**
  * Creates the handler for validating GitHub issues against the codebase.
  *
- * Uses Claude SDK with:
+ * Uses the provider abstraction with:
  * - Read-only tools (Read, Glob, Grep) for codebase analysis
  * - JSON schema structured output for reliable parsing
  * - System prompt guiding the validation process
@@ -229,6 +261,7 @@ export function createValidateIssueHandler(
         issueBody,
         issueLabels,
         model = 'opus',
+        thinkingLevel,
         comments: rawComments,
         linkedPRs: rawLinkedPRs,
       } = req.body as ValidateIssueRequestBody;
@@ -276,11 +309,14 @@ export function createValidateIssueHandler(
         return;
       }
 
-      // Validate model parameter at runtime
-      if (!VALID_MODELS.includes(model)) {
+      // Validate model parameter at runtime - accept Claude models or Cursor models
+      const isValidClaudeModel = VALID_CLAUDE_MODELS.includes(model as ModelAlias);
+      const isValidCursorModel = isCursorModel(model);
+
+      if (!isValidClaudeModel && !isValidCursorModel) {
         res.status(400).json({
           success: false,
-          error: `Invalid model. Must be one of: ${VALID_MODELS.join(', ')}`,
+          error: `Invalid model. Must be one of: ${VALID_CLAUDE_MODELS.join(', ')}, or a Cursor model ID`,
         });
         return;
       }
@@ -310,7 +346,8 @@ export function createValidateIssueHandler(
         abortController,
         settingsService,
         validationComments,
-        validationLinkedPRs
+        validationLinkedPRs,
+        thinkingLevel
       )
         .catch(() => {
           // Error is already handled inside runValidation (event emitted)

apps/server/src/routes/ideation/common.ts

@@ -0,0 +1,12 @@
+/**
+ * Common utilities for ideation routes
+ */
+
+import { createLogger } from '@automaker/utils';
+import { getErrorMessage as getErrorMessageShared, createLogError } from '../common.js';
+
+const logger = createLogger('Ideation');
+
+// Re-export shared utilities
+export { getErrorMessageShared as getErrorMessage };
+export const logError = createLogError(logger);

apps/server/src/routes/ideation/index.ts

@@ -0,0 +1,109 @@
+/**
+ * Ideation routes - HTTP API for brainstorming and idea management
+ */
+
+import { Router } from 'express';
+import type { EventEmitter } from '../../lib/events.js';
+import { validatePathParams } from '../../middleware/validate-paths.js';
+import type { IdeationService } from '../../services/ideation-service.js';
+import type { FeatureLoader } from '../../services/feature-loader.js';
+
+// Route handlers
+import { createSessionStartHandler } from './routes/session-start.js';
+import { createSessionMessageHandler } from './routes/session-message.js';
+import { createSessionStopHandler } from './routes/session-stop.js';
+import { createSessionGetHandler } from './routes/session-get.js';
+import { createIdeasListHandler } from './routes/ideas-list.js';
+import { createIdeasCreateHandler } from './routes/ideas-create.js';
+import { createIdeasGetHandler } from './routes/ideas-get.js';
+import { createIdeasUpdateHandler } from './routes/ideas-update.js';
+import { createIdeasDeleteHandler } from './routes/ideas-delete.js';
+import { createAnalyzeHandler, createGetAnalysisHandler } from './routes/analyze.js';
+import { createConvertHandler } from './routes/convert.js';
+import { createAddSuggestionHandler } from './routes/add-suggestion.js';
+import { createPromptsHandler, createPromptsByCategoryHandler } from './routes/prompts.js';
+import { createSuggestionsGenerateHandler } from './routes/suggestions-generate.js';
+
+export function createIdeationRoutes(
+  events: EventEmitter,
+  ideationService: IdeationService,
+  featureLoader: FeatureLoader
+): Router {
+  const router = Router();
+
+  // Session management
+  router.post(
+    '/session/start',
+    validatePathParams('projectPath'),
+    createSessionStartHandler(ideationService)
+  );
+  router.post('/session/message', createSessionMessageHandler(ideationService));
+  router.post('/session/stop', createSessionStopHandler(events, ideationService));
+  router.post(
+    '/session/get',
+    validatePathParams('projectPath'),
+    createSessionGetHandler(ideationService)
+  );
+
+  // Ideas CRUD
+  router.post(
+    '/ideas/list',
+    validatePathParams('projectPath'),
+    createIdeasListHandler(ideationService)
+  );
+  router.post(
+    '/ideas/create',
+    validatePathParams('projectPath'),
+    createIdeasCreateHandler(events, ideationService)
+  );
+  router.post(
+    '/ideas/get',
+    validatePathParams('projectPath'),
+    createIdeasGetHandler(ideationService)
+  );
+  router.post(
+    '/ideas/update',
+    validatePathParams('projectPath'),
+    createIdeasUpdateHandler(events, ideationService)
+  );
+  router.post(
+    '/ideas/delete',
+    validatePathParams('projectPath'),
+    createIdeasDeleteHandler(events, ideationService)
+  );
+
+  // Project analysis
+  router.post('/analyze', validatePathParams('projectPath'), createAnalyzeHandler(ideationService));
+  router.post(
+    '/analysis',
+    validatePathParams('projectPath'),
+    createGetAnalysisHandler(ideationService)
+  );
+
+  // Convert to feature
+  router.post(
+    '/convert',
+    validatePathParams('projectPath'),
+    createConvertHandler(events, ideationService, featureLoader)
+  );
+
+  // Add suggestion to board as a feature
+  router.post(
+    '/add-suggestion',
+    validatePathParams('projectPath'),
+    createAddSuggestionHandler(ideationService, featureLoader)
+  );
+
+  // Guided prompts (no validation needed - static data)
+  router.get('/prompts', createPromptsHandler(ideationService));
+  router.get('/prompts/:category', createPromptsByCategoryHandler(ideationService));
+
+  // Generate suggestions (structured output)
+  router.post(
+    '/suggestions/generate',
+    validatePathParams('projectPath'),
+    createSuggestionsGenerateHandler(ideationService)
+  );
+
+  return router;
+}

apps/server/src/routes/ideation/routes/add-suggestion.ts

@@ -0,0 +1,70 @@
+/**
+ * POST /add-suggestion - Add an analysis suggestion to the board as a feature
+ *
+ * This endpoint converts an AnalysisSuggestion to a Feature using the
+ * IdeationService's mapIdeaCategoryToFeatureCategory for consistent category mapping.
+ * This ensures a single source of truth for the conversion logic.
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import type { FeatureLoader } from '../../../services/feature-loader.js';
+import type { AnalysisSuggestion } from '@automaker/types';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createAddSuggestionHandler(
+  ideationService: IdeationService,
+  featureLoader: FeatureLoader
+) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, suggestion } = req.body as {
+        projectPath: string;
+        suggestion: AnalysisSuggestion;
+      };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      if (!suggestion) {
+        res.status(400).json({ success: false, error: 'suggestion is required' });
+        return;
+      }
+
+      if (!suggestion.title) {
+        res.status(400).json({ success: false, error: 'suggestion.title is required' });
+        return;
+      }
+
+      if (!suggestion.category) {
+        res.status(400).json({ success: false, error: 'suggestion.category is required' });
+        return;
+      }
+
+      // Build description with rationale if provided
+      const description = suggestion.rationale
+        ? `${suggestion.description}\n\n**Rationale:** ${suggestion.rationale}`
+        : suggestion.description;
+
+      // Use the service's category mapping for consistency
+      const featureCategory = ideationService.mapSuggestionCategoryToFeatureCategory(
+        suggestion.category
+      );
+
+      // Create the feature
+      const feature = await featureLoader.create(projectPath, {
+        title: suggestion.title,
+        description,
+        category: featureCategory,
+        status: 'backlog',
+      });
+
+      res.json({ success: true, featureId: feature.id });
+    } catch (error) {
+      logError(error, 'Add suggestion to board failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/analyze.ts

@@ -0,0 +1,49 @@
+/**
+ * POST /analyze - Analyze project and generate suggestions
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createAnalyzeHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath } = req.body as { projectPath: string };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      // Start analysis - results come via WebSocket events
+      ideationService.analyzeProject(projectPath).catch((error) => {
+        logError(error, 'Analyze project failed (async)');
+      });
+
+      res.json({ success: true, message: 'Analysis started' });
+    } catch (error) {
+      logError(error, 'Analyze project failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}
+
+export function createGetAnalysisHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath } = req.body as { projectPath: string };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      const result = await ideationService.getCachedAnalysis(projectPath);
+      res.json({ success: true, result });
+    } catch (error) {
+      logError(error, 'Get analysis failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/convert.ts

@@ -0,0 +1,77 @@
+/**
+ * POST /convert - Convert an idea to a feature
+ */
+
+import type { Request, Response } from 'express';
+import type { EventEmitter } from '../../../lib/events.js';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import type { FeatureLoader } from '../../../services/feature-loader.js';
+import type { ConvertToFeatureOptions } from '@automaker/types';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createConvertHandler(
+  events: EventEmitter,
+  ideationService: IdeationService,
+  featureLoader: FeatureLoader
+) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, ideaId, keepIdea, column, dependencies, tags } = req.body as {
+        projectPath: string;
+        ideaId: string;
+      } & ConvertToFeatureOptions;
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      if (!ideaId) {
+        res.status(400).json({ success: false, error: 'ideaId is required' });
+        return;
+      }
+
+      // Convert idea to feature structure
+      const featureData = await ideationService.convertToFeature(projectPath, ideaId);
+
+      // Apply any options from the request
+      if (column) {
+        featureData.status = column;
+      }
+      if (dependencies && dependencies.length > 0) {
+        featureData.dependencies = dependencies;
+      }
+      if (tags && tags.length > 0) {
+        featureData.tags = tags;
+      }
+
+      // Create the feature using FeatureLoader
+      const feature = await featureLoader.create(projectPath, featureData);
+
+      // Delete the idea unless keepIdea is explicitly true
+      if (!keepIdea) {
+        await ideationService.deleteIdea(projectPath, ideaId);
+
+        // Emit idea deleted event
+        events.emit('ideation:idea-deleted', {
+          projectPath,
+          ideaId,
+        });
+      }
+
+      // Emit idea converted event to notify frontend
+      events.emit('ideation:idea-converted', {
+        projectPath,
+        ideaId,
+        featureId: feature.id,
+        keepIdea: !!keepIdea,
+      });
+
+      // Return featureId as expected by the frontend API interface
+      res.json({ success: true, featureId: feature.id });
+    } catch (error) {
+      logError(error, 'Convert to feature failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/ideas-create.ts

@@ -0,0 +1,51 @@
+/**
+ * POST /ideas/create - Create a new idea
+ */
+
+import type { Request, Response } from 'express';
+import type { EventEmitter } from '../../../lib/events.js';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import type { CreateIdeaInput } from '@automaker/types';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createIdeasCreateHandler(events: EventEmitter, ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, idea } = req.body as {
+        projectPath: string;
+        idea: CreateIdeaInput;
+      };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      if (!idea) {
+        res.status(400).json({ success: false, error: 'idea is required' });
+        return;
+      }
+
+      if (!idea.title || !idea.description || !idea.category) {
+        res.status(400).json({
+          success: false,
+          error: 'idea must have title, description, and category',
+        });
+        return;
+      }
+
+      const created = await ideationService.createIdea(projectPath, idea);
+
+      // Emit idea created event for frontend notification
+      events.emit('ideation:idea-created', {
+        projectPath,
+        idea: created,
+      });
+
+      res.json({ success: true, idea: created });
+    } catch (error) {
+      logError(error, 'Create idea failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/ideas-delete.ts

@@ -0,0 +1,42 @@
+/**
+ * POST /ideas/delete - Delete an idea
+ */
+
+import type { Request, Response } from 'express';
+import type { EventEmitter } from '../../../lib/events.js';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createIdeasDeleteHandler(events: EventEmitter, ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, ideaId } = req.body as {
+        projectPath: string;
+        ideaId: string;
+      };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      if (!ideaId) {
+        res.status(400).json({ success: false, error: 'ideaId is required' });
+        return;
+      }
+
+      await ideationService.deleteIdea(projectPath, ideaId);
+
+      // Emit idea deleted event for frontend notification
+      events.emit('ideation:idea-deleted', {
+        projectPath,
+        ideaId,
+      });
+
+      res.json({ success: true });
+    } catch (error) {
+      logError(error, 'Delete idea failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/ideas-get.ts

@@ -0,0 +1,39 @@
+/**
+ * POST /ideas/get - Get a single idea
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createIdeasGetHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, ideaId } = req.body as {
+        projectPath: string;
+        ideaId: string;
+      };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      if (!ideaId) {
+        res.status(400).json({ success: false, error: 'ideaId is required' });
+        return;
+      }
+
+      const idea = await ideationService.getIdea(projectPath, ideaId);
+      if (!idea) {
+        res.status(404).json({ success: false, error: 'Idea not found' });
+        return;
+      }
+
+      res.json({ success: true, idea });
+    } catch (error) {
+      logError(error, 'Get idea failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/ideas-list.ts

@@ -0,0 +1,26 @@
+/**
+ * POST /ideas/list - List all ideas for a project
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createIdeasListHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath } = req.body as { projectPath: string };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      const ideas = await ideationService.getIdeas(projectPath);
+      res.json({ success: true, ideas });
+    } catch (error) {
+      logError(error, 'List ideas failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/ideas-update.ts

@@ -0,0 +1,54 @@
+/**
+ * POST /ideas/update - Update an idea
+ */
+
+import type { Request, Response } from 'express';
+import type { EventEmitter } from '../../../lib/events.js';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import type { UpdateIdeaInput } from '@automaker/types';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createIdeasUpdateHandler(events: EventEmitter, ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, ideaId, updates } = req.body as {
+        projectPath: string;
+        ideaId: string;
+        updates: UpdateIdeaInput;
+      };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      if (!ideaId) {
+        res.status(400).json({ success: false, error: 'ideaId is required' });
+        return;
+      }
+
+      if (!updates) {
+        res.status(400).json({ success: false, error: 'updates is required' });
+        return;
+      }
+
+      const idea = await ideationService.updateIdea(projectPath, ideaId, updates);
+      if (!idea) {
+        res.status(404).json({ success: false, error: 'Idea not found' });
+        return;
+      }
+
+      // Emit idea updated event for frontend notification
+      events.emit('ideation:idea-updated', {
+        projectPath,
+        ideaId,
+        idea,
+      });
+
+      res.json({ success: true, idea });
+    } catch (error) {
+      logError(error, 'Update idea failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/prompts.ts

@@ -0,0 +1,42 @@
+/**
+ * GET /prompts - Get all guided prompts
+ * GET /prompts/:category - Get prompts for a specific category
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import type { IdeaCategory } from '@automaker/types';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createPromptsHandler(ideationService: IdeationService) {
+  return async (_req: Request, res: Response): Promise<void> => {
+    try {
+      const prompts = ideationService.getAllPrompts();
+      const categories = ideationService.getPromptCategories();
+      res.json({ success: true, prompts, categories });
+    } catch (error) {
+      logError(error, 'Get prompts failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}
+
+export function createPromptsByCategoryHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { category } = req.params as { category: string };
+
+      const validCategories = ideationService.getPromptCategories().map((c) => c.id);
+      if (!validCategories.includes(category as IdeaCategory)) {
+        res.status(400).json({ success: false, error: 'Invalid category' });
+        return;
+      }
+
+      const prompts = ideationService.getPromptsByCategory(category as IdeaCategory);
+      res.json({ success: true, prompts });
+    } catch (error) {
+      logError(error, 'Get prompts by category failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/session-get.ts

@@ -0,0 +1,45 @@
+/**
+ * POST /session/get - Get an ideation session with messages
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createSessionGetHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, sessionId } = req.body as {
+        projectPath: string;
+        sessionId: string;
+      };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      if (!sessionId) {
+        res.status(400).json({ success: false, error: 'sessionId is required' });
+        return;
+      }
+
+      const session = await ideationService.getSession(projectPath, sessionId);
+      if (!session) {
+        res.status(404).json({ success: false, error: 'Session not found' });
+        return;
+      }
+
+      const isRunning = ideationService.isSessionRunning(sessionId);
+
+      res.json({
+        success: true,
+        session: { ...session, isRunning },
+        messages: session.messages,
+      });
+    } catch (error) {
+      logError(error, 'Get session failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/session-message.ts

@@ -0,0 +1,40 @@
+/**
+ * POST /session/message - Send a message in an ideation session
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import type { SendMessageOptions } from '@automaker/types';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createSessionMessageHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { sessionId, message, options } = req.body as {
+        sessionId: string;
+        message: string;
+        options?: SendMessageOptions;
+      };
+
+      if (!sessionId) {
+        res.status(400).json({ success: false, error: 'sessionId is required' });
+        return;
+      }
+
+      if (!message) {
+        res.status(400).json({ success: false, error: 'message is required' });
+        return;
+      }
+
+      // This is async but we don't await - responses come via WebSocket
+      ideationService.sendMessage(sessionId, message, options).catch((error) => {
+        logError(error, 'Send message failed (async)');
+      });
+
+      res.json({ success: true });
+    } catch (error) {
+      logError(error, 'Send message failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/session-start.ts

@@ -0,0 +1,30 @@
+/**
+ * POST /session/start - Start a new ideation session
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import type { StartSessionOptions } from '@automaker/types';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createSessionStartHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, options } = req.body as {
+        projectPath: string;
+        options?: StartSessionOptions;
+      };
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      const session = await ideationService.startSession(projectPath, options);
+      res.json({ success: true, session });
+    } catch (error) {
+      logError(error, 'Start session failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/session-stop.ts

@@ -0,0 +1,39 @@
+/**
+ * POST /session/stop - Stop an ideation session
+ */
+
+import type { Request, Response } from 'express';
+import type { EventEmitter } from '../../../lib/events.js';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import { getErrorMessage, logError } from '../common.js';
+
+export function createSessionStopHandler(events: EventEmitter, ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { sessionId, projectPath } = req.body as {
+        sessionId: string;
+        projectPath?: string;
+      };
+
+      if (!sessionId) {
+        res.status(400).json({ success: false, error: 'sessionId is required' });
+        return;
+      }
+
+      await ideationService.stopSession(sessionId);
+
+      // Emit session stopped event for frontend notification
+      // Note: The service also emits 'ideation:session-ended' internally,
+      // but we emit here as well for route-level consistency with other routes
+      events.emit('ideation:session-ended', {
+        sessionId,
+        projectPath,
+      });
+
+      res.json({ success: true });
+    } catch (error) {
+      logError(error, 'Stop session failed');
+      res.status(500).json({ success: false, error: getErrorMessage(error) });
+    }
+  };
+}

apps/server/src/routes/ideation/routes/suggestions-generate.ts

@@ -0,0 +1,56 @@
+/**
+ * Generate suggestions route - Returns structured AI suggestions for a prompt
+ */
+
+import type { Request, Response } from 'express';
+import type { IdeationService } from '../../../services/ideation-service.js';
+import { createLogger } from '@automaker/utils';
+import { getErrorMessage, logError } from '../common.js';
+
+const logger = createLogger('ideation:suggestions-generate');
+
+export function createSuggestionsGenerateHandler(ideationService: IdeationService) {
+  return async (req: Request, res: Response): Promise<void> => {
+    try {
+      const { projectPath, promptId, category, count } = req.body;
+
+      if (!projectPath) {
+        res.status(400).json({ success: false, error: 'projectPath is required' });
+        return;
+      }
+
+      if (!promptId) {
+        res.status(400).json({ success: false, error: 'promptId is required' });
+        return;
+      }
+
+      if (!category) {
+        res.status(400).json({ success: false, error: 'category is required' });
+        return;
+      }
+
+      // Default to 10 suggestions, allow 1-20
+      const suggestionCount = Math.min(Math.max(count || 10, 1), 20);
+
+      logger.info(`Generating ${suggestionCount} suggestions for prompt: ${promptId}`);
+
+      const suggestions = await ideationService.generateSuggestions(
+        projectPath,
+        promptId,
+        category,
+        suggestionCount
+      );
+
+      res.json({
+        success: true,
+        suggestions,
+      });
+    } catch (error) {
+      logError(error, 'Failed to generate suggestions');
+      res.status(500).json({
+        success: false,
+        error: getErrorMessage(error),
+      });
+    }
+  };
+}

apps/server/src/routes/mcp/common.ts

@@ -2,6 +2,10 @@
  * Common utilities for MCP routes
  */
 
+import { createLogger } from '@automaker/utils';
+
+const logger = createLogger('MCP');
+
 /**
  * Extract error message from unknown error
  */
@@ -16,5 +20,5 @@ export function getErrorMessage(error: unknown): string {
  * Log error with prefix
  */
 export function logError(error: unknown, message: string): void {
-  console.error(`[MCP] ${message}:`, error);
+  logger.error(`${message}:`, error);
 }

apps/server/src/routes/models/routes/available.ts

@@ -1,61 +1,16 @@
 /**
- * GET /available endpoint - Get available models
+ * GET /available endpoint - Get available models from all providers
  */
 
 import type { Request, Response } from 'express';
+import { ProviderFactory } from '../../../providers/provider-factory.js';
 import { getErrorMessage, logError } from '../common.js';
 
-interface ModelDefinition {
-  id: string;
-  name: string;
-  provider: string;
-  contextWindow: number;
-  maxOutputTokens: number;
-  supportsVision: boolean;
-  supportsTools: boolean;
-}
-
 export function createAvailableHandler() {
   return async (_req: Request, res: Response): Promise<void> => {
     try {
-      const models: ModelDefinition[] = [
-        {
-          id: 'claude-opus-4-5-20251101',
-          name: 'Claude Opus 4.5',
-          provider: 'anthropic',
-          contextWindow: 200000,
-          maxOutputTokens: 16384,
-          supportsVision: true,
-          supportsTools: true,
-        },
-        {
-          id: 'claude-sonnet-4-20250514',
-          name: 'Claude Sonnet 4',
-          provider: 'anthropic',
-          contextWindow: 200000,
-          maxOutputTokens: 16384,
-          supportsVision: true,
-          supportsTools: true,
-        },
-        {
-          id: 'claude-3-5-sonnet-20241022',
-          name: 'Claude 3.5 Sonnet',
-          provider: 'anthropic',
-          contextWindow: 200000,
-          maxOutputTokens: 8192,
-          supportsVision: true,
-          supportsTools: true,
-        },
-        {
-          id: 'claude-3-5-haiku-20241022',
-          name: 'Claude 3.5 Haiku',
-          provider: 'anthropic',
-          contextWindow: 200000,
-          maxOutputTokens: 8192,
-          supportsVision: true,
-          supportsTools: true,
-        },
-      ];
+      // Get all models from all registered providers (Claude + Cursor)
+      const models = ProviderFactory.getAllAvailableModels();
 
       res.json({ success: true, models });
     } catch (error) {

apps/server/src/routes/models/routes/providers.ts

@@ -17,6 +17,13 @@ export function createProvidersHandler() {
           available: statuses.claude?.installed || false,
           hasApiKey: !!process.env.ANTHROPIC_API_KEY,
         },
+        cursor: {
+          available: statuses.cursor?.installed || false,
+          version: statuses.cursor?.version,
+          path: statuses.cursor?.path,
+          method: statuses.cursor?.method,
+          authenticated: statuses.cursor?.authenticated,
+        },
       };
 
       res.json({ success: true, providers });