- Introduced comprehensive unit tests for exec-utils, covering execAsync, extendedPath, execEnv, and isENOENT functions.
- Added tests for FeatureVerificationService to validate feature verification and commit processes.
- Implemented tests for ProjectAnalyzer to ensure project analysis functionality.
- Created tests for TaskExecutor to validate task execution and event emissions.
These additions enhance test coverage and ensure the reliability of core functionalities.
- Update feature-loader.ts to use logger.info instead of console.log
- Update dev-server-service.ts to use createLogger for all logging
- Update terminal-service.ts to use createLogger for all logging
- Update agent-service.ts to use createLogger for all logging
- Fix agent-service.test.ts mock to preserve actual createLogger
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Updated getPending method to ensure that only features with a defined status are included in the pending features list.
- Improved code readability by using a more concise filter function.
This change enhances the robustness of feature status handling.
- Added isENOENT function to routes/common.ts for improved error handling.
- Re-exported isENOENT alongside existing utilities for better accessibility.
This change streamlines error management in file system operations.
- Update 26+ files to import secureFs from @automaker/platform
- Create shared type files: github.ts, worktree.ts, claude.ts in libs/types
- Create exec-utils.ts for cross-platform shell execution
- Delete redundant wrapper files: secure-fs.ts, stream-processor.ts, enhancement-prompts.ts
- Update GitHub routes to use createLogError pattern
- Add isENOENT helper to routes/common.ts
- Fix test imports to use @automaker/prompts and @automaker/platform
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Reduce auto-mode-service.ts from 1308 to 516 lines (60% reduction) by
extracting reusable functionality into shared packages and services:
- Add feature prompt builders to @automaker/prompts (buildFeaturePrompt,
buildFollowUpPrompt, buildContinuationPrompt, extractTitleFromDescription)
- Add planning prompts and task parsing to @automaker/prompts
- Add stream processor utilities to @automaker/utils (sleep, processStream)
- Add git commit utilities to @automaker/git-utils (commitAll, hasUncommittedChanges)
- Create ProjectAnalyzer service for project analysis
- Create FeatureVerificationService for verify/commit operations
- Extend FeatureLoader with updateStatus, updatePlanSpec, getPending methods
- Expand FeatureStatus type to include all used statuses
- Add PlanSpec and ParsedTask types to @automaker/types
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Introduced a new workflow in release.yml to automate the release process for macOS, Windows, and Linux.
- Added a script (update-version.mjs) to update the version in package.json based on the release tag.
- Configured artifact uploads for each platform and ensured proper version extraction and validation.
- Added rehype-sanitize as a dependency to sanitize Markdown content.
- Updated the Markdown component to include rehype-sanitize in the rehypePlugins for improved security against XSS attacks.
- Updated the list-issues and list-prs handlers to fetch open and closed issues, as well as open and merged PRs in parallel, improving performance.
- Removed the redundant 'issues' and 'prs' properties from the result interfaces to streamline the response structure.
- Added 'skipTests' flag in integration tests to indicate tests that should be skipped, enhancing test management.
- Introduced a comprehensive E2E Testing Guide outlining best practices for Playwright tests, including principles for test isolation, element selection, and setup utilities.
- Added new test files for project creation and opening existing projects, ensuring functionality for creating blank projects and projects from GitHub templates.
- Implemented utility functions for setting up test states and managing localStorage, enhancing maintainability and reducing boilerplate in tests.
Add platform detection to apply additional left padding (pl-20) and top
padding (pt-4) on macOS to prevent the sidebar header/logo from
overlapping with the native window control buttons (close, minimize,
maximize).
Fixes#176
- Add internal _dev:* scripts without build:packages prefix
- Update dev:full to call build:packages once, then use internal scripts via concurrently
- This prevents build:packages from running 3 times (once in dev:full, once in dev:server, once in dev:web)
- Keep build scripts simple with direct approach (no duplication issue to solve)
Addresses gemini-code-assist bot feedback on PR #215🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Add build:packages to prepare hook for automatic builds after npm install
- Prefix all dev:* scripts with build:packages to ensure packages are built before development
- Prefix all build:* scripts with build:packages to ensure packages are built before production builds
This ensures developers never encounter "module not found" errors from unbuilt packages in libs/ directory.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Modified the build command to first execute the build:packages script, ensuring all necessary packages are built before the UI.
- Streamlined the build:packages command by consolidating workspace flags for better readability and maintenance.
- Added a mock for the `loadContextFiles` function to return an empty context by default, improving test reliability.
- Updated the agent-service test suite to ensure proper initialization of the `AgentService` with mocked dependencies.
These changes aim to enhance the test coverage and stability of the agent-service functionality.
- Implemented path validation against ALLOWED_ROOT_DIRECTORY to prevent arbitrary file reads and prompt injection attacks.
- Added error handling for file reading, including specific responses for forbidden paths and file not found scenarios.
- Updated the description generation logic to truncate large files and provide structured prompts for analysis.
- Enhanced logging for better traceability of file access and errors.
These changes aim to improve the security and reliability of the file description functionality.
- Introduced a new utility function `loadContextFiles` to load project-specific context files from the `.automaker/context/` directory, enhancing agent prompts with project rules and guidelines.
- Updated `AgentService` and `AutoModeService` to utilize the new context loading functionality, combining context prompts with existing system prompts for improved agent performance.
- Added comprehensive documentation on the context files system, including usage examples and metadata structure, to facilitate better understanding and implementation.
- Removed redundant context loading logic from `AutoModeService`, streamlining the codebase.
These changes aim to improve the agent's contextual awareness and adherence to project-specific conventions.
