Remove confusing and unnecessary delete calls from resumeFeature() and
resumePipelineFeature() methods. These were leftovers from a previous
implementation where temporary entries were added to runningFeatures.
The resumeFeature() method already ensures the feature is not running
at the start (via has() check that throws if already running), so these
delete calls serve no purpose and only add confusion about state management.
Removed delete calls from:
- resumeFeature() non-pipeline flow (line 748)
- resumePipelineFeature() no-context case (line 798)
- resumePipelineFeature() step-not-found case (line 822)
Co-authored-by: gemini-code-assist bot
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Reduce redundant file reads by loading the feature object once and passing
it down the call chain instead of reloading it multiple times.
Changes:
- Pass feature object to resumePipelineFeature() instead of featureId
- Pass feature object to resumeFromPipelineStep() instead of featureId
- Remove redundant loadFeature() calls from these methods
- Add FeatureStatusWithPipeline import for type safety
This improves performance by eliminating unnecessary file I/O operations
and makes the data flow clearer.
Co-authored-by: gemini-code-assist bot
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Added support for grouped models in the PhaseModelSelector, allowing users to select from multiple variants within a single group.
- Introduced a new popover UI for displaying grouped model variants, improving user interaction and selection clarity.
- Implemented logic to filter and display enabled cursor models, including standalone and grouped options.
- Enhanced state management for expanded groups and variant selection, ensuring a smoother user experience.
This update significantly improves the model selection process, making it more intuitive and organized.
This commit fixes several edge cases when resuming features stuck in pipeline steps:
- Detect if feature is stuck in a pipeline step during resume
- Handle case where context file is missing (restart from beginning)
- Handle case where pipeline step no longer exists in config
- Add dedicated resumePipelineFeature() method for pipeline-specific resume logic
- Add detectPipelineStatus() to extract and validate pipeline step information
- Add resumeFromPipelineStep() to resume from a specific pipeline step index
- Update board view to check context availability for features with pipeline status
Edge cases handled:
1. No context file → restart entire pipeline from beginning
2. Step no longer exists in config → complete feature without pipeline
3. Valid step exists → resume from the crashed step
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Added defaultExpanded prop to TaskProgressPanel for customizable initial state.
- Updated styling in TaskProgressPanel for improved layout and consistency.
- Modified AgentOutputModal to utilize the new defaultExpanded prop and adjusted class names for better responsiveness and appearance.
- Enhanced overflow handling in AgentOutputModal for improved user experience.
- Introduced skipSandboxWarning property in GlobalSettings interface to manage user preference for sandbox risk warnings.
- Updated syncSettingsToServer function to include skipSandboxWarning in the settings synchronization process.
- Set default value for skipSandboxWarning to false in DEFAULT_GLOBAL_SETTINGS.
- Updated the SandboxRiskDialog to include a checkbox for users to opt-out of future warnings, passing the state to the onConfirm callback.
- Modified SettingsView to manage the skipSandboxWarning state, allowing users to reset the warning preference.
- Enhanced DangerZoneSection to display a message when the sandbox warning is disabled and provide an option to reset this setting.
- Updated RootLayoutContent to respect the user's choice regarding the sandbox warning, auto-confirming if the user opts to skip it.
- Added skipSandboxWarning state management to the app store for persistent user preferences.
- Introduced a popover for model selection, allowing users to choose from Claude and Cursor models.
- Added functionality to toggle favorite models, enhancing user experience by allowing quick access to preferred options.
- Updated the UI to display selected model details and improved layout for better usability.
- Refactored model grouping and rendering logic for clarity and maintainability.
This update improves the overall interaction with the phase model selector, making it more intuitive and user-friendly.
- Added a new auth store using Zustand to manage authentication state, including `authChecked` and `isAuthenticated`.
- Updated `LoginView` to set authentication state upon successful login and navigate based on setup completion.
- Enhanced `RootLayoutContent` to enforce routing rules based on authentication status, redirecting users to login or setup as necessary.
- Improved error handling and loading states during authentication checks.
- Add handler for type=result messages in Cursor stream processing
- Cursor provider sends final accumulated text in msg.result
- Suggestions was only handling assistant messages for Cursor
- Add detailed logging for result extraction (like backlog plan)
- Matches pattern used by github validation and backlog plan
This fixes potential parsing issues when using Cursor models
for feature suggestions, ensuring the complete response text
is captured before JSON parsing.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Removed the 'tier' property from Cursor model configurations and related UI components.
- Updated relevant files to reflect the removal of tier-related logic and display elements.
- This change simplifies the model structure and UI, focusing on essential attributes.
- Import isCursorModel to detect Cursor models
- For Cursor: embed systemPrompt in userPrompt with explicit instructions
- Add "DO NOT write any files" directive for Cursor models
- Prevents Cursor from writing plan to files instead of returning JSON
- Matches pattern used by github validation (validate-issue.ts)
Cursor doesn't support systemPrompt separation like Claude SDK,
so we need to combine prompts and add explicit instructions to
prevent it from using Write/Edit tools and creating files.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Change debug logs to info/warn so they're always visible
- Log when result message is received from Cursor
- Log lengths of both msg.result and accumulated responseText
- Log which source is being used (result vs accumulated)
- Log empty response error for better diagnostics
- Add response preview logging on parse failure
This will help diagnose why Cursor parsing is failing by showing:
1. Whether result messages are being received
2. What content lengths we're working with
3. Whether response text is empty or has content
4. What the actual response looks like when parsing fails
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Switch from extractJson to extractJsonWithArray for 'changes' field
- Validates that 'changes' is an array, not just that it exists
- Add debug logging for response length and preview on parse failure
- Add debug logging when receiving result from Cursor provider
- Matches pattern used by suggestions feature
- Helps diagnose parsing issues with better error context
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Add handler for type=result messages in stream processing
- Cursor provider sends final accumulated text in msg.result
- Backlog plan was only handling assistant messages
- Now matches pattern used by github validation and suggestions
- Fixes "cursor cli parsing failed" error
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Add ModelOverrideTrigger to backlog plan dialog
- Position trigger in DialogFooter on left side (mr-auto)
- Display before Cancel button for better UX
- Use variant="button" to show model name
- Connect to phaseModels.backlogPlanningModel default
- Pass model override to server generate endpoint
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Added ModelOverrideTrigger component to the "Plan Backlog Changes" dialog,
allowing users to override the global backlog planning model on a per-request
basis, consistent with other dialogs in the application.
Changes:
- Added model override state management to backlog-plan-dialog
- Integrated ModelOverrideTrigger component in dialog header (input mode only)
- Pass model override (or global default) to backlogPlan.generate API call
- UI shows override indicator when model is overridden from global default
The feature uses the existing backlogPlanningModel phase setting as the default
and allows temporary overrides without changing global settings.
Server already supports optional model parameter in the generate endpoint,
so no backend changes were required.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Fixed cloud storage path detection to work correctly on Windows by normalizing
path separators to forward slashes and removing Windows drive letters before
pattern matching.
Issue:
The isCloudStoragePath() function was failing on Windows because:
1. path.resolve() converts Unix paths to Windows paths with backslashes
2. Windows adds drive letters (e.g., "P:\Users\test" instead of "/Users/test")
3. Pattern checks for "/Library/CloudStorage/" didn't match "\Library\CloudStorage\"
4. Home-anchored path comparisons failed due to drive letter mismatches
Solution:
- Normalize all path separators to forward slashes for consistent pattern matching
- Remove Windows drive letters (e.g., "C:" or "P:") from normalized paths
- This ensures Unix-style test paths work the same on all platforms
All tests now pass (967 passed, 27 skipped):
- ✅ Cloud storage path detection tests (macOS patterns)
- ✅ Home-anchored cloud folder tests (Dropbox, Google Drive, OneDrive)
- ✅ Sandbox compatibility tests
- ✅ Cross-platform path handling
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Updated terminal-service.test.ts to use path.resolve() in test expectations
so they work correctly on both Unix and Windows platforms.
The merge from main removed the skipIf conditions for Windows, expecting these
tests to work cross-platform. On Windows, path.resolve('/test/dir') converts
Unix-style paths to Windows paths (e.g., 'P:\test\dir'), so test expectations
needed to use path.resolve() as well to match the actual behavior.
Fixed tests:
- should create a new terminal session
- should fix double slashes in path
- should return all active sessions
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Carefully merged latest changes from main branch into the Cursor CLI integration
branch. This merge brings in important improvements and fixes while preserving
all Cursor-related functionality.
Key changes from main:
- Sandbox mode security improvements and cloud storage compatibility
- Version-based settings migrations (v2 schema)
- Port configuration centralization
- System paths utilities for CLI detection
- Enhanced error handling in HttpApiClient
- Windows MCP process cleanup fixes
- New validation and build commands
- GitHub issue templates and release process improvements
Resolved conflicts in:
- apps/server/src/routes/context/routes/describe-image.ts
(Combined Cursor provider routing with secure-fs imports)
- apps/server/src/services/auto-mode-service.ts
(Merged failure tracking with raw output logging)
- apps/server/tests/unit/services/terminal-service.test.ts
(Updated to async tests with systemPathExists mocking)
- libs/platform/src/index.ts
(Combined WSL utilities with system-paths exports)
- libs/types/src/settings.ts
(Merged DEFAULT_PHASE_MODELS with SETTINGS_VERSION constants)
All Cursor CLI integration features remain intact including:
- CursorProvider and CliProvider base class
- Phase-based model configuration
- Provider registry and factory patterns
- WSL support for Windows
- Model override UI components
- Cursor-specific settings and configurations
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- apps/server/src/providers/claude-provider.ts
Add ANTHROPIC_BASE_URL and ANTHROPIC_AUTH_TOKEN to the environment variable
allowlist, enabling use of LLM gateways (LiteLLM, Helicone) and Anthropic-
compatible providers (GLM 4.7, Minimax M2.1, etc.).
Closes#338
- Updated the checkSandboxCompatibility function to explicitly handle the case when enableSandboxMode is set to false, ensuring clearer logic for sandbox mode activation.
- Adjusted unit tests to reflect the new behavior, confirming that sandbox mode defaults to enabled when not specified and correctly disables for cloud storage paths.
- Enhanced test descriptions for better clarity on expected outcomes in various scenarios.
- Introduced macOS-specific cloud storage patterns and home-anchored folder detection to improve accuracy in identifying cloud storage paths.
- Updated the isCloudStoragePath function to utilize these new patterns, ensuring better handling of cloud storage locations.
- Added comprehensive unit tests to validate detection logic for various cloud storage scenarios, including false positive prevention.
- Added error handling for HTTP responses in the HttpApiClient class.
- Enhanced error messages to include status text and parsed error data, improving debugging and user feedback.
- Modified the application to keep the app and servers running when all windows are closed on macOS, aligning with standard macOS behavior.
- On other platforms, ensured that the server processes are stopped and the app quits when all windows are closed, preventing potential port conflicts.
- Added a new function to check if a port is in use without terminating processes, improving user experience during server startup.
- Updated the health check function to accept a dynamic port parameter, allowing for flexible server configurations.
- Implemented user prompts for handling port conflicts, enabling users to kill processes, choose different ports, or cancel the operation.
- Enhanced CORS configuration to support localhost and IPv6 addresses, ensuring compatibility across different development environments.
- Refactored the main function to utilize dynamic port assignments for both the web and server applications, improving overall flexibility.
- Updated the bumpVersion function to use plain JavaScript by removing TypeScript type annotations, improving compatibility with non-TypeScript environments.
- Cleaned up whitespace in the bump-version.mjs file for better readability.
- Added versioning to global settings, enabling automatic migrations for breaking changes.
- Updated default global settings to reflect the new versioning schema.
- Implemented logic to disable sandbox mode for existing users during migration from version 1 to 2.
- Enhanced error handling for saving migrated settings, ensuring data integrity during updates.
- Introduced a new command for fetching and validating GitHub issues, allowing users to address issues directly from the command line.
- Added a release command to bump the version of the application and build the Electron app, ensuring version consistency across UI and server packages.
- Updated package.json files for both UI and server to version 0.7.1, reflecting the latest changes.
- Implemented version utility in the server to read the version from package.json, enhancing version management across the application.
- Replaced the existing PNG logo with a new SVG version for improved scalability and quality.
- Added the SVG logo file to the project, enhancing visual consistency across different display resolutions.
- Updated test repository creation functions to utilize environment variables for git author and committer information, preventing modifications to the user's global git configuration.
- This change enhances test isolation and ensures consistent behavior across different environments.
- Added an environment variable to simulate a containerized environment, allowing the application to skip sandbox confirmation dialogs during testing.
- This change aims to streamline the testing process by reducing unnecessary user interactions while ensuring the application behaves as expected in a containerized setup.
- Introduced a new endpoint to check if the application is running in a containerized environment, allowing the UI to display appropriate risk warnings.
- Added a confirmation dialog for users when running outside a sandbox, requiring acknowledgment of potential risks before proceeding.
- Implemented a rejection screen for users who deny sandbox risk confirmation, providing options to restart in a container or reload the application.
- Updated the main application logic to handle sandbox status checks and user responses effectively, enhancing security and user experience.
- Introduced a new command for validating tests, providing detailed instructions for running tests and fixing failures based on code changes.
- Updated the environment variable handling in the Claude provider to only allow explicitly defined variables, enhancing security and preventing leakage of sensitive information.
- Improved feature loading to handle errors more gracefully and load features concurrently, optimizing performance.
- Centralized port configuration for the Automaker application to prevent accidental termination of critical services.
- Start API key initialization immediately upon importing the HTTP API client module to ensure the init promise is created early.
- Log errors during API key initialization to aid in debugging.
Additionally, added a version field to the setup store for proper state hydration, aligning with the app-store pattern.
API requests were being made before initApiKey() completed, causing
401 Unauthorized errors on app startup in Electron mode.
Changes:
- Add waitForApiKeyInit() to track and await API key initialization
- Make HTTP methods (get/post/put/delete) wait for auth before requests
- Defer WebSocket connection until API key is ready
- Add explicit auth wait in useSettingsMigration hook
Fixes race condition introduced in PR #321
- Updated the TerminalService to utilize getShellPaths() for better shell detection across platforms.
- Improved logic for detecting user-configured shells in WSL and added fallbacks for various platforms.
- Enhanced unit tests to mock shell paths for comprehensive cross-platform testing, ensuring accurate shell detection behavior.
These changes aim to streamline shell detection and improve the user experience across different operating systems.
- Introduced a .nvmrc file to specify the Node.js version (22) for the project, ensuring consistent development environments.
- Enhanced error handling in the startServer function to provide clearer messages when the Node.js executable cannot be found, improving debugging experience.
- Updated package.json files across various modules to enforce Node.js version compatibility and ensure consistent dependency versions.
These changes aim to streamline development processes and enhance the application's reliability by enforcing version control and improving error reporting.
- Introduced a restricted file system wrapper to ensure all file operations are confined to the script's directory, enhancing security.
- Updated various modules to utilize the new secure file system methods, replacing direct fs calls with validated operations.
- Enhanced path validation in the server routes and context loaders to prevent unauthorized access to the file system.
- Adjusted environment variable handling to use centralized methods for reading and writing API keys, ensuring consistent security practices.
This change improves the overall security posture of the application by enforcing strict file access controls and validating paths before any operations are performed.
- Refactor agent-view.tsx from 1028 lines to ~215 lines
- Create agent-view/ folder with components/, hooks/, input-area/, shared/
- Extract hooks: useAgentScroll, useFileAttachments, useAgentShortcuts, useAgentSession
- Extract components: AgentHeader, ChatArea, MessageList, MessageBubble, ThinkingIndicator
- Extract input-area: AgentInputArea, FilePreview, QueueDisplay, InputControls
- Add AgentModelSelector with Claude and Cursor CLI model support
- Update /models/available to use ProviderFactory.getAllAvailableModels()
- Update /models/providers to include Cursor CLI status
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
