This commit fixes bidirectional data synchronization between Electron and Web
modes by addressing multiple interconnected issues:
**Core Fixes:**
1. **Electron userData Path (main.ts)**
- Explicitly set userData path in development using app.setPath()
- Navigate from __dirname to project root instead of relying on process.cwd()
- Ensures Electron reads from /data instead of ~/.config/Automaker
2. **Server DataDir Path (main.ts, start-automaker.sh)**
- Fixed startServer() to use __dirname for reliable path calculation
- Export DATA_DIR environment variable in start-automaker.sh
- Server now consistently uses shared /data directory
3. **Settings Sync Protection (settings-service.ts)**
- Modified wipe protection to distinguish legitimate removals from accidents
- Allow empty projects array if trashedProjects has items
- Prevent false-positive wipe detection when removing projects
4. **Diagnostics & Logging**
- Enhanced cache loading logging in use-settings-migration.ts
- Detailed migration decision logs for troubleshooting
- Track project counts from both cache and server
**Impact:**
- Projects created in Electron now appear in Web mode after restart
- Projects removed in Web mode stay removed in Electron after restart
- Settings changes sync bidirectionally across mode switches
- No more data loss or project duplication issues
**Testing:**
- Verified Electron uses /home/dhanush/Projects/automaker/data
- Confirmed server startup logs show correct DATA_DIR
- Tested project persistence across mode restarts
- Validated no writes to ~/.config/Automaker in dev mode
Fixes: Data persistence between Electron and Web modes
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
Updated the PR state handling to use a consistent uppercase format ('OPEN', 'MERGED', 'CLOSED') throughout the codebase. This includes changes to the worktree metadata interface, PR creation logic, and related tests to ensure uniformity and prevent potential mismatches in state representation.
Additionally, modified the GitHub PR fetching logic to retrieve all PR states, allowing for better detection of state changes.
This refactor enhances clarity and consistency in how PR states are managed and displayed.
The web mode launcher was setting CORS_ORIGIN to only include the system
hostname and 127.0.0.1, but users access via http://localhost:3007 which
wasn't in the allowed list.
Now includes:
- http://localhost:3007 (primary dev URL)
- http://$HOSTNAME:3007 (system hostname if needed)
- http://127.0.0.1:3007 (loopback IP)
Also cleaned up debug logging from CORS check since root cause is now clear.
Fixes: Persistent "Not allowed by CORS" errors in web mode
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
Added detailed logging to see:
- What origin is being sent
- How the hostname is parsed
- Why origins are being accepted/rejected
This will help us understand why CORS is still failing in web mode.
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
The CORS check was too strict for local development. Changed to:
- Parse origin URL properly to extract hostname
- Allow all localhost origins (any port)
- Allow all 127.0.0.1 origins (loopback IP)
- Allow all private network IPs (192.168.x.x, 10.x.x.x, 172.x.x.x)
- Keep security by rejecting unknown origins
This fixes CORS errors when accessing from http://localhost:3007
or other local addresses during web mode development.
Fixes: "Not allowed by CORS" errors in web mode
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
Fix race condition where sandbox warning appeared on every refresh
even after checking "Do not show again". The issue was that the
sandbox check effect ran before settings were hydrated from the
server, so skipSandboxWarning was always false (the default).
Changes:
- Add settingsLoaded to sandbox check dependencies to ensure the
user's preference is loaded before checking
- Add AUTOMAKER_SKIP_SANDBOX_WARNING env var option to skip the
warning entirely (useful for dev/CI environments)
When the dev server restarts, developers need to re-enter the API key
in the browser. While the key is persisted to ./data/.api-key, this
file may be missing in clean dev scenarios.
This adds a helpful tip to the API key banner informing developers
they can set AUTOMAKER_API_KEY environment variable for a persistent
API key during development, avoiding the need to re-enter it after
server restarts.
When opening a git repository without a GitHub remote, the server logs
were spammed with warnings every 5 seconds during worktree polling:
WARN [Worktree] Failed to fetch GitHub PRs: Command failed: gh pr list
... no git remotes found
This happened because fetchGitHubPRs() ran `gh pr list` without first
checking if the project has a GitHub remote configured.
Changes:
- Add per-project cache for GitHub remote status with 5-minute TTL
- Check cache before attempting to fetch PRs, skip silently if no remote
- Add forceRefreshGitHub parameter to clear cache on manual refresh
- Pass forceRefreshGitHub when user clicks the refresh worktrees button
This allows users to add a GitHub remote and immediately detect it by
clicking the refresh button, while preventing log spam during normal
polling for projects without GitHub remotes.
Resolved conflict in backlog-plan/common.ts:
- Kept local (stricter) validation: Array.isArray(parsed?.result?.changes)
- This ensures type safety for the changes array
- Added a new `/sync` endpoint to synchronize the project specification with the current codebase and feature state.
- Introduced `syncSpec` function to handle the synchronization logic, updating technology stack, implemented features, and roadmap phases.
- Enhanced the running state management to track synchronization tasks alongside existing generation tasks.
- Updated UI components to support synchronization actions, including loading indicators and status updates.
- Improved logging and error handling for better visibility during sync operations.
These changes enhance project management capabilities by ensuring that the specification remains up-to-date with the latest code and feature developments.
- Introduced `HOSTNAME` environment variable for user-facing URLs, defaulting to localhost.
- Updated server and client code to utilize `HOSTNAME` for constructing URLs instead of hardcoded localhost.
- Enhanced documentation in CLAUDE.md to reflect new configuration options.
- Added `VITE_HOSTNAME` for frontend API URLs, ensuring consistent hostname usage across the application.
These changes improve flexibility in server configuration and enhance the user experience by providing accurate URLs.
- Add HOST environment variable (default: 0.0.0.0) to allow binding to specific network interfaces
- Update server to listen on configurable host instead of hardcoded localhost
- Update Vite dev server to respect HOST environment variable
- Enhanced server startup banner to display listening address
- Updated .env.example and CLAUDE.md documentation
Fixes#536
Co-authored-by: Web Dev Cody <webdevcody@users.noreply.github.com>
- Removed redundant definition of CLI base timeout in `cli-provider.ts` and added a detailed comment explaining its purpose.
- Updated `codex-provider.ts` to use the imported `DEFAULT_TIMEOUT_MS` directly instead of an alias.
- Enhanced unit tests to ensure fallback behavior for invalid reasoning effort values in timeout calculations.
- Introduced atomic write functionality for JSON files to ensure data integrity during writes.
- Added recovery mechanisms to read JSON files with fallback options for corrupted or missing files.
- Enhanced existing services to utilize atomic write and recovery features for improved reliability.
- Updated tests to cover new atomic writing and recovery scenarios, ensuring robust error handling and data consistency.
- Added `calculateReasoningTimeout` function to dynamically adjust timeouts based on reasoning effort levels.
- Updated CLI and Codex providers to utilize the new timeout calculation, addressing potential timeouts for high reasoning efforts.
- Enhanced unit tests to validate timeout behavior for various reasoning efforts, ensuring correct timeout values are applied.
- Added Notification Service to manage project-level notifications, including creation, listing, marking as read, and dismissing notifications.
- Introduced Event History Service to store and manage historical events, allowing for listing, retrieval, deletion, and replaying of events.
- Integrated notifications into the server and UI, providing real-time updates for feature statuses and operations.
- Enhanced sidebar and project switcher components to display unread notifications count.
- Created dedicated views for managing notifications and event history, improving user experience and accessibility.
These changes enhance the application's ability to inform users about important events and statuses, improving overall usability and responsiveness.
- Updated test to send a specific authentication error pattern to the data callback.
- Triggered the exit handler to validate the handling of authentication errors.
- Improved error message expectations for better clarity during test failures.
- Introduced a new xml-extractor module with functions for XML parsing, including escaping/unescaping XML characters, extracting sections and elements, and managing implemented features.
- Added functionality to add, remove, update, and check for implemented features in the app_spec.txt file.
- Enhanced the create and update feature handlers to check for duplicate titles and trigger synchronization with app_spec.txt on status changes.
- Updated tests to cover new XML extraction utilities and feature handling logic, ensuring robust functionality and reliability.
- Updated the version in package.json for the main project to 0.12.0rc.
- Updated the version in apps/server/package.json and apps/ui/package.json to 0.12.0.
- Adjusted the version extraction logic in start-automaker.sh to reference the correct package.json path.
- Added detection for Electron environment to improve compatibility with Windows PTY processes.
- Implemented winpty fallback for ConPTY failures, ensuring robust terminal session creation in Electron and other contexts.
- Updated error handling to provide clearer messages for authentication and terminal access issues.
- Refined usage data detection logic to avoid false positives, improving the accuracy of usage reporting.
These changes aim to enhance the reliability and user experience of terminal interactions on Windows, particularly in Electron applications.
Added a mock for the Unix platform in the SIGTERM test case to ensure proper behavior during testing on non-Windows systems. This change enhances the reliability of the tests by simulating the expected environment for process termination.
Refactored the process termination logic in both ClaudeUsageService and TerminalService to use a centralized method for killing PTY processes. This ensures consistent handling of process termination across Windows and Unix-like systems, improving reliability and maintainability of the code.
Updated the process termination logic in ClaudeUsageService to handle Windows environments correctly. The code now checks the operating system and calls the appropriate kill method, ensuring consistent behavior across platforms.
Updated the JSON parsing in the loadBacklogPlan function to explicitly cast the raw input as a string, improving type safety and preventing potential runtime errors when handling backlog plan data.
- Simplify escapeXml() using 'str == null' check (type narrowing)
- Add validation for extracted JSON before passing to specToXml()
- Prevents runtime errors when JSON doesn't match SpecOutput schema
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Ensure running details are cleared when generation completes or fails, preventing state leaks.
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
- Add setRunningDetails(null) in finally block of generate handler to prevent state leaks
- Move clearBacklogPlan before response in apply handler and wrap in try-catch to prevent errors after headers sent
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
Documentation improvements:
- Fix GitHub URL placeholder issues in install-fedora.md - GitHub /latest/download/ endpoint
doesn't support version substitution, use explicit download URL pattern instead
- Improve security in network troubleshooting section:
- Change ping target from claude.ai (marketing site) to api.anthropic.com (actual API)
- Remove unsafe 'echo \$ANTHROPIC_API_KEY' command that exposes secrets in shell history
- Use safe API key check with conditional output instead
Code improvements:
- apps/server/src/routes/backlog-plan/common.ts: Add Array.isArray() validation
for stored plan shape before returning it. Ensures changes is actually an array,
not just truthy, preventing downstream runtime errors.
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
Fix type mismatch in loadBacklogPlan where secureFs.readFile with 'utf-8'
encoding returns union type string | Buffer, causing JSON.parse to fail type checking.
Cast raw to string to satisfy TypeScript strict mode.
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
Fixes spec generation failure when using custom API endpoints (e.g., GLM proxy)
that don't support structured output. The AI returns JSON instead of XML, but
the fallback parser only looked for XML tags.
Changes:
- escapeXml: Handle undefined/null values gracefully (converts to empty string)
- generate-spec: Add JSON extraction fallback when XML tags aren't found
- Reuses existing extractJson() utility (already used for Cursor models)
- Converts extracted JSON to XML using specToXml()
Closes#510🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Introduced server log level configuration and HTTP request logging settings, allowing users to control the verbosity of server logs and enable or disable request logging at runtime.
- Added an Event Hook Service to execute custom actions based on system events, supporting shell commands and HTTP webhooks.
- Enhanced the UI with new sections for managing server logging preferences and event hooks, including a dialog for creating and editing hooks.
- Updated global settings to include server log level and request logging options, ensuring persistence across sessions.
These changes aim to improve debugging capabilities and provide users with customizable event-driven actions within the application.
- Added functionality to save, clear, and load backlog plans within the application.
- Introduced a new API endpoint for clearing saved backlog plans.
- Enhanced the backlog plan dialog to allow users to review and apply changes to their features.
- Integrated dependency management features in the UI, allowing users to select parent and child dependencies for features.
- Improved the graph view with options to manage plans and visualize dependencies effectively.
- Updated the sidebar and settings to include provider visibility toggles for better user control over model selection.
These changes aim to enhance the user experience by providing robust backlog management capabilities and improving the overall UI for feature planning.
- auto-mode-service-planning.test.ts: Add taskExecutionPrompts argument
to buildFeaturePrompt calls, update test for implementation instructions
- claude-usage-service.test.ts: Skip deprecated Mac tests (service now
uses PTY for all platforms), rename Windows tests to PTY tests, update
to use process.cwd() instead of home directory
- claude-provider.test.ts: Add missing model parameter to environment
variable passthrough tests
All tests now pass (1093 passed, 23 skipped).
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Update auto-mode-service.ts to use centralized Task Execution prompts
from settings, making all 9 task execution prompts customizable via UI:
- buildFeaturePrompt: uses implementationInstructions and
playwrightVerificationInstructions from settings
- buildTaskPrompt: uses taskPromptTemplate with variable substitution
- buildPipelineStepPrompt: updated to pass prompts through
- executeFeatureWithContext: uses resumeFeatureTemplate
- resolvePlanApproval recovery: uses continuationAfterApprovalTemplate
- Multi-agent continuation: uses continuationAfterApprovalTemplate
- recordLearningsFromFeature: uses learningExtractionSystemPrompt
and learningExtractionUserPromptTemplate
All 12 prompt categories are now fully customizable from the UI.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Fix inline type imports in defaults.ts (move to top-level imports)
- Update ideation-service.ts to use centralized prompts from settings
- Update generate-title.ts to use centralized prompts
- Update validate-issue.ts to use centralized prompts
- Clean up validation-schema.ts (prompts already centralized)
- Minor server index cleanup
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add 4 new prompt type interfaces (AppSpecPrompts, ContextDescriptionPrompts,
SuggestionsPrompts, TaskExecutionPrompts) with resolved types
- Add default prompts for all new categories to @automaker/prompts/defaults.ts
- Add merge functions for new prompt categories in merge.ts
- Update settings-helpers.ts getPromptCustomization() to return all 12 categories
- Update server routes (generate-spec, generate-features-from-spec, describe-file,
describe-image, generate-suggestions) to use centralized prompts
- Add 4 new tabs in prompt customization UI (App Spec, Context, Suggestions, Tasks)
- Fix Ideation tab layout using grid-cols-4 for even distribution
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Added functionality for bulk verifying features in the BoardView, allowing users to mark multiple features as verified at once.
- Introduced a selection target mechanism to differentiate between 'backlog' and 'waiting_approval' features during selection mode.
- Updated the KanbanCard and SelectionActionBar components to support the new selection target logic, improving user experience for bulk actions.
- Enhanced the UI to provide appropriate actions based on the current selection target, including verification options for waiting approval features.
