- Shareable links now load faster, with improved caching and routing speed! ⚡
- New authentication flow simplifies sign-in, handling edge cases smoothly! 🔐
- Refined UI components deliver cleaner layouts and more consistent styling! 🎨
- Expanded API endpoints add powerful capabilities for integrations and automation! 🔌
- Background jobs are more reliable, with better retries and observability! 🛠️
- Database migrations improved for safer deploys and faster rollbacks! 🗄️
- Enhanced error handling provides clearer messages and fewer silent failures! 🚨
- Updated dependency versions boost security, stability, and overall performance! 🧩
- Better logging and metrics make debugging production issues much easier! 📊
- Improved documentation clarifies setup steps and common troubleshooting paths! 📚
Fixes#75 - Linux ARM64 builds were failing because:
1. Missing build-essential (gcc, g++, make) needed to compile node-pty
2. Missing Python 3.11 for node-gyp (Python 3.12+ removed distutils)
Also adds architecture verification step that fails early if node-pty
is incorrectly built for x86_64 instead of ARM64.
Fixes#67 - Claude Code not detected when installed via default macOS installer.
Two issues were causing agent detection failures:
1. Tilde (~) in custom paths wasn't being expanded before filesystem checks.
Users entering ~/.local/bin/claude as a custom path would fail because
Node.js fs functions don't understand shell tilde expansion.
2. macOS lacked direct path probing for common installation locations.
Unlike Windows which had probeWindowsPaths(), macOS relied solely on
'which' which doesn't find binaries outside PATH or shell aliases.
Changes:
- Add expandTilde() helper to convert ~ to os.homedir()
- Apply tilde expansion in checkCustomPath() before all fs operations
- Add probeUnixPaths() method to directly check known installation paths:
- ~/.claude/local/claude (default Claude installer location)
- ~/.local/bin/claude
- /opt/homebrew/bin/claude (Homebrew Apple Silicon)
- /usr/local/bin/claude (Homebrew Intel)
- And equivalent paths for codex, opencode, gemini, aider
- Integrate probeUnixPaths() into checkBinaryExists() as first-pass detection
- AICommandsPanel: add expandCommand helper and update 19 tests for
new collapsible command UI (buttons only visible when expanded)
- AgentSessionsBrowser/AgentSessionsModal: update tool display text
assertions to match ToolCallCard format (without brackets)
- SettingsModal: add SpecKitCommandsPanel mock and tabShortcuts props
- Share usable project context faster with smarter, consolidated workspace insights! 🚀
- Enjoy cleaner navigation through streamlined structure and simplified module organization! 🧭
- Experience improved stability from refreshed dependencies and safer version upgrades! 🛡️
- Get better performance through targeted optimizations and reduced redundant processing! ⚡
- Debug issues quicker with enhanced logging, clearer errors, and richer diagnostics! 🔍
- Ship confidently with upgraded CI workflows and more reliable automated checks! ✅
- Benefit from tighter security with improved secrets handling and safer defaults! 🔐
- Configure setups easier thanks to more flexible options and clearer configuration paths! 🧩
- Developers move faster with improved docs, examples, and onboarding instructions! 📚
- Reduce maintenance headaches with cleaner code patterns and smarter abstractions! 🧰
- Share and manage projects faster with improved GitHub repository organization! 🚀
- Enjoy smoother collaboration through upgraded workflows and cleaner update tracking! ✨
- Get clearer release visibility with more consistent, polished changelog updates! 📝
Send the GitHub compare link (or commit/release notes) since last release, and I’ll generate precise 10-word bullets. 🔎
The OpenCode image test was hardcoded to use 'ollama/qwen3-vl'
which failed when ollama wasn't configured. Now it:
- Respects OPENCODE_VISION_MODEL environment variable
- Falls back to ollama/qwen3-vl if not set
- Skips the image test when OPENCODE_VISION_MODEL is not configured
This allows running OpenCode tests with different vision models
(e.g., github-copilot/gemini-3-flash-preview) or skipping the
image test entirely when no vision-capable model is available.
- Share builds instantly with web and mobile playback improvements across devices 🚀
- Enjoy faster startup times thanks to core runtime performance optimizations ⚡
- Customize workflows with new CLI flags and smarter default configuration behavior 🛠️
- Upgrade dependency stack for improved security, stability, and compatibility 🔒
- Simplify CI pipelines with refined GitHub Actions and caching enhancements 🧩
- Improve error messages and logs for quicker debugging and clearer diagnosis 🔎
- Add better TypeScript support with stricter typings and updated build outputs ✨
- Enhance plugin system with new hooks and more reliable extension loading 🔌
- Strengthen tests with broader coverage and more deterministic test execution 🧪
- Refresh documentation with clearer guides, examples, and updated release notes 📚
- Share credential definitions, plugins, templates, and workflows across organizations now! 🔗
- Expanded credential management: create, import, export, and move seamlessly! 🚀
- Faster, clearer organization switching with improved UX and navigation flow! 🧭
- New API endpoints for org-shared credentials, templates, and workflows added! 🧩
- Enhanced permissions and access controls for shared resources across orgs! 🔒
- Improved audit logging and traceability for credential and workflow operations! 🧾
- UI updates for managing shared assets: cleaner tables and actions! ✨
- Better error handling and validation for imports and credential operations! 🛠️
- Performance improvements in list views, searches, and organization loading! ⚡
- Documentation refreshed with new sharing features and updated integration guides! 📚
Add ToolCallCard component that displays tool execution details from
session history with collapsible sections for better readability.
Features:
- Collapsible card header showing [Tool: name] with status indicator
- Expandable Input section with JSON formatting
- Expandable Output section with JSON formatting
- Time and Status fields
- Show more/less links for long content
- Supports both Claude (name) and OpenCode (tool) tool formats
Files:
- ToolCallCard.tsx: New reusable component for tool display
- AgentSessionsBrowser.tsx: Integrate ToolCallCard, add getToolName helper
- AgentSessionsModal.tsx: Integrate ToolCallCard for consistency
The drag-and-drop copy feature (hold Cmd/Ctrl while dragging) was not
working reliably due to several issues:
1. handleDragLeave was clearing dropTargetIndex before the drop could
occur, causing race conditions where the drop handler found null state
2. The browser's drop event doesn't always fire reliably (e.g., when
cursor is slightly outside container boundaries)
3. handleDragOver wasn't properly validating positions for move vs copy
mode
Changes:
- Remove state clearing from handleDragLeave (cleanup now only happens
in handleDrop and handleDragEnd)
- Add dropPerformedRef to prevent double execution if both handleDrop
and handleDragEnd fire
- Add fallback logic in handleDragEnd to perform the operation if
handleDrop never fired
- Improve handleDragOver to properly validate positions (always show
indicato
The drag-and-drop copy feature (hold Cmd/Ctrl while dragging) was not
working reliably due to severagit reset HEAD~1 2>/dev/null; git add src/renderer/components/DocumentsPanel.tsx
cd /Users/oliveiraantoniocc/Documents/projects/Maestro
'
- Commands now display collapsed by default with chevron toggle
- Expanded view shows Edit/Delete buttons and truncated prompt preview
- Edit mode inline with Cancel/Save buttons
- Consistent styling with Spec Kit Commands panel
- Better space efficiency for long command lists
- Share and publish your project easier with improved release workflows! 🚀
- Enjoy faster performance from major dependency and build optimizations! ⚡
- Get clearer docs and examples for smoother setup and onboarding! 📚
- Strengthen reliability with expanded tests and improved CI coverage! ✅
- Fix critical bugs across core features for more stable runs! 🛠️
- Upgrade APIs and interfaces for cleaner integration and extension! 🔌
- Enhance developer experience with better tooling and linting updates! 🧰
- Improve configuration options for more flexible, powerful deployments! 🎛️
- Refine UI/UX interactions for a more polished experience overall! ✨
- Tighten security and maintenance with refreshed packages and patches! 🔒
- Share custom GPTs publicly with a new dedicated store experience! 🚀
- Browse and discover community-built GPTs faster with improved search! 🔎
- Install and manage GPTs easily with clearer listings and details! 🧩
- New monetization options help creators earn from popular GPTs! 💸
- Better safety reviews and policy enforcement for shared GPTs! 🛡️
- Enhanced admin controls for organizations to govern GPT access! 🏢
- Improved analytics for creators to track usage and engagement! 📈
- Updated UI streamlines building, testing, and publishing GPTs quicker! ✨
- Expanded integrations and actions make GPTs more useful everyday! 🔌
- Performance and reliability improvements across the GPT platform overall! ⚡
- Add clickable action links to toast notifications (for PR URLs)
- Add history entry when PR is created with branch info and description
- Show uncommitted changes warning in CreatePRModal before creating PR
- Add DeleteWorktreeModal with three options: Cancel, Confirm (keep files), Confirm and Delete on Disk
- Add git:removeWorktree IPC handler to delete worktree directories from disk
- Share your current tab with friends instantly using a simple link! 🔗
- Broadcast your voice alongside browsing for richer hangout sessions! 🎙️
- Enjoy smoother, faster syncing with major performance improvements throughout! ⚡
- Experience a refreshed UI that’s cleaner, clearer, and easier to use! ✨
- Control privacy better with improved permissions and safer defaults now! 🛡️
- Fix annoying desync bugs and stay perfectly in-step together! 🧩
- Upgrade reliability with stronger reconnect handling and session recovery! 🔄
- Support more sites and edge cases with expanded compatibility updates! 🌍
- Simplify setup with improved onboarding and clearer configuration flows! 🚀
- Polish the overall experience with numerous stability fixes and refinements! 🧼
- Register spec-kit IPC handlers in main process (was missing)
- Implement actual download/extract for "Check for Updates" button
- Downloaded prompts stored in user data dir, preferred over bundled
- Add spec-kit prompts to extraResources for production builds
- Fix github/spec-kit link to open in system browser
- Add /speckit.help command explaining workflow and Auto Run integration
- Simplify metadata.json (remove redundant commands object)
- Resolved conflicts in package.json (kept both test:performance and refresh-speckit scripts)
- Resolved conflicts in agent-detector.test.ts (kept try/finally pattern for platform restoration)
- Fixed speckit-manager.ts to read prompts from disk instead of using Vite's ?raw imports
- Removed unused extractDescription function and added proper types for GitHub API response
- Add ESLint 9 configuration (eslint.config.mjs) with TypeScript and React hooks plugins
- Add npm run lint:eslint command for code quality checks
- Expand npm run lint to check all three TypeScript configs (renderer, main, cli)
- Update tsconfig.cli.json to include src/prompts and src/types directories
Fix 29 ESLint errors:
- Remove unused updateCliActivity import in batch-processor.ts
- Convert {false && <jsx>} patterns to comments in AutoRun components
- Wrap case block with const declarations in braces (AgentSelectionScreen)
- Fix unused expression pattern in PreparingPlanScreen
- Fix conditional hook calls in FilePreview, OfflineQueueBanner, RecentCommandChips, TerminalOutput
- Add windows-diagnostics.json to PackageContents interface
Update CLAUDE.md and CONTRIBUTING.md with new linting commands and documentation.
Claude ID: 029e8abe-5734-4967-9fb4-c85078c1973d
Maestro ID: 87ffa06e-0ecd-4eb8-b327-dad1ec24f7a9
Created 29 tests covering all 7 playbooks handlers:
- playbooks:list - 4 tests for listing playbooks
- playbooks:create - 4 tests for creating with ID generation
- playbooks:update - 3 tests for updating existing playbooks
- playbooks:delete - 2 tests for single playbook deletion
- playbooks:deleteAll - 3 tests for session cleanup
- playbooks:export - 5 tests for ZIP export functionality
- playbooks:import - 8 tests for ZIP import with manifest parsing
Tests include proper mocking for electron dialog, fs/promises,
archiver, adm-zip, and crypto modules following the established
test patterns from agentSessions.test.ts.
Add 35 tests covering all 8 persistence IPC handlers:
- settings:get - retrieves setting by key, handles missing/nested keys
- settings:set - stores values, broadcasts theme/commands to web clients
- settings:getAll - returns all settings from store
- sessions:getAll - loads sessions from store
- sessions:setAll - writes sessions, detects and broadcasts changes
- groups:getAll - loads groups from store
- groups:setAll - writes groups to store
- cli:getActivity - reads CLI activity file, handles errors gracefully
Add test file with 26 tests covering all process IPC handlers:
- process:spawn: Tests for PTY spawn, pid return, failure handling, env vars
- process:write: Tests for stdin writing, invalid session handling
- process:kill: Tests for process termination
- process:interrupt: Tests for SIGINT signal sending
- process:resize: Tests for PTY dimension resizing
- process:getActiveProcesses: Tests for listing running processes
- process:runCommand: Tests for command execution with custom shells
All tests use the established pattern from agentSessions.test.ts with
proper mocking of electron ipcMain, ProcessManager, AgentDetector, and stores.
- Created 70 tests covering all 23 system-related IPC handlers
- Tests cover dialog, fonts, shells, tunnel, devtools, updates, logger, and sync handlers
- Follows established testing pattern from agentSessions.test.ts
- All tests pass including integration with existing test suite (410 total tests)
Added 15 error handling tests covering:
- File permission errors (EACCES) for listSessions, listSessionsPaginated,
readSessionMessages, deleteMessagePair, and searchSessions
- Disk full errors (ENOSPC) for deleteMessagePair with proper error propagation
- Network path unavailable errors (ENOENT, ETIMEDOUT, EHOSTUNREACH,
ECONNREFUSED, EIO) across various handlers
- Combined error scenarios with mixed readable/unreadable sessions
All 74 tests now pass.
Added 11 tests covering:
- Delete message pair by UUID
- Error handling when user message not found
- Fallback content matching when UUID match fails
- Multiple assistant messages deletion until next user message
- End-of-file deletion with no subsequent user message
- Array content handling for fallback matching
- Orphaned tool_result cleanup after tool_use deletion
- Malformed JSON line handling
- Missing session file error
- Message preservation before/after deleted pair
- Empty file handling
Also marked 3 task items as N/A since the handlers don't exist:
- claude:deleteSession (no such handler)
- claude:getSessionPath (no such handler)
- claude:getStoragePath (no such handler)
- Added 10 tests covering session listing functionality:
- Returns sessions from ~/.claude directory with correct parsing
- Returns empty array when project directory does not exist
- Filters out 0-byte session files
- Parses session JSON files and extracts token counts
- Adds origin info from origins store (object and string formats)
- Extracts first user message text from array content (handles images)
- Sorts sessions by modified date descending
- Handles malformed JSON lines gracefully
- Calculates cost estimate from token counts
Enhanced the handler registration test to document all 14 ipcMain.handle calls
with their source line numbers and brief descriptions for easier maintenance.
