- Add automatic DataCache cleanup with 30-second intervals
- Implement size limit enforcement to prevent memory buildup
- Increase memory threshold to 300MB for realistic analytics usage
- Add Claude session tracking integration throughout dashboard
- Fix message counting to respect actual Claude session windows
- Stabilize conversation state transitions with extended time intervals
- Add comprehensive token breakdown with cache tokens
- Implement Claude session popover with hover functionality
- Fix table layout stability with fixed column widths
- Reduce cache size limits from 1000 to 100 entries
- Add proper cleanup methods for graceful shutdown
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fix token calculation discrepancy by including cache tokens in main dashboard
- Add detailed token breakdown popover with hover functionality
- Improve conversation history loading with comprehensive message metadata
- Fix message coloring logic using JSON metadata instead of content parsing
- Add version number integration to dashboard
- Enhance SessionTimer with info icon and message calculation tooltip
- Move notifications button to filter row for better UX
- Add comprehensive token usage tracking (input, output, cache creation, cache read)
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Implement comprehensive unit tests for StateCalculator to validate conversation status detection and state management.
- Create unit tests for StateService covering state management, subscriber notifications, and error handling.
- Develop unit tests for WebSocketServer to ensure proper handling of WebSocket connections, message broadcasting, and heartbeat mechanisms.
- Include edge case handling and validation for all components to enhance robustness and reliability.
- Implement 500ms refresh rate for near real-time updates
- Add 250ms fast refresh for first 30 seconds for instant state changes
- Prioritize conversation flow over file timestamps for immediate response detection
- Improve "Claude Code working..." detection to trigger instantly when user sends message
- Extend file activity detection window to 15 seconds for better accuracy
- Create dedicated npm README for better package discoverability
- Update root README to highlight new notification and state detection features
- Remove debug console logs for cleaner production experience
Performance improvements result in near-instantaneous state change detection.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Implement real-time conversation state updates with 1-second refresh
- Add precise state detection: "Claude Code working...", "User typing...", "Awaiting user input..."
- Separate refresh logic: manual for charts, automatic for conversation data
- Integrate active process detection for accurate state determination
- Improve notification system for awaiting user input states
- Remove emojis from notification titles and refresh buttons for cleaner interface
- Optimize performance by only refreshing conversation data, not charts
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Remove emojis from notification and refresh buttons
- Clean, professional button text appearance
- Maintain all functionality while improving visual clarity
- Better accessibility and readability
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Implement browser notification API integration
- Add notification button to enable/disable alerts
- Detect conversation state changes (Awaiting user input...)
- Send notifications when Claude is waiting for user response
- Include clickable notifications that focus the conversation
- Auto-close notifications after 10 seconds
- Visual button states (enabled/disabled/denied)
- Test notification on first enable
- Improve productivity by alerting users when Claude needs input
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Change default selection from 'setup' to 'analytics'
- Promote analytics dashboard usage as primary feature
- Users can still easily select Project Setup if needed
- Improves discoverability of monitoring capabilities
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Remove automatic help output when no arguments provided
- Allow interactive welcome screen to show cleanly
- Users can still access help with --help flag
- Fixes clean onboarding experience
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add initial prompt to choose between Analytics Dashboard or Project Setup
- Show clear descriptions for each option with emojis
- Only show prompt when no specific CLI flags are provided
- Maintain existing behavior for users with flags (--yes, --language, etc.)
- Improve onboarding experience for new users
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Bump version to 1.5.18
- Sync web dashboard HTML with latest analytics improvements
- Add process indicators and orphan process handling UI elements
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Refine conversation status detection to reduce false positives
- Filter out system processes and orphan process noise
- Balance active conversation detection for better accuracy
- Only show truly active conversations in dashboard
- Remove orphan process detection to reduce UI clutter
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Bump version in package.json from 1.5.12 to 1.5.14.
- Add CSS styles for improved layout and status indicators in the web dashboard.
- Introduce a new message type indicator for user and assistant messages with tooltips.
- Implement logic to determine message types (success, error, tool execution, pending) based on message content.
- Refactor message rendering to include message type indicators.
- Updated onboarding prompts to include an option for launching the Claude Code Analytics Dashboard.
- Modified the createClaudeConfig function to handle the analytics option and initiate the dashboard if selected by the user.
- Added dynamic column width calculation based on actual command names
- Implemented visual separators (│) between columns for better readability
- Command names are capped at 25 characters with ellipsis truncation
- Responsive table width that adapts to content
- Better alignment and spacing for all columns
- Cleaner, more professional table appearance
- Fixes deformed tables when command names are long
Example improvement:
Before: command-name-is-very-long1.2KB 25 180 350 7/9/2025
After: command-name-is-very-l... │ 1.2KB │ 25 │ 180 │ 350 │ 7/9/2025
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Changed 'Tokens' column header to 'Tokens (aprox)' for clarity
- Added promptSetupWhenNoCommands() function to offer setup when no commands exist
- Updated table width to accommodate longer header (97 characters)
- When no .claude/commands directory is found, offers to start Claude Code Templates setup
- Improved user experience by providing clear next steps
- Returns appropriate boolean values from displayCommandStats for flow control
- Seamless integration between command analysis and initial setup
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Added estimateTokens() function for token estimation based on content analysis
- Considers markdown formatting, code blocks, and text complexity
- New 'Tokens' column in command analysis table with magenta color
- Updated table width and formatting to accommodate token data
- Shows total token count in summary (e.g., '~3,833 tokens total')
- Helps users understand command complexity in terms of Claude token usage
- Essential for managing prompt lengths and optimization decisions
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Added --command-stats CLI option to analyze existing Claude Code commands
- Created command-stats.js module with comprehensive analysis functions
- Displays command statistics in formatted table (name, size, lines, words, last modified)
- Calculates totals and provides project overview
- Interactive prompt for command optimization using Claude Code
- Generates intelligent optimization prompts based on project structure
- Integrates with existing CLI workflow and error handling
- Helps users optimize their .claude/commands for their specific project setup
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Removed duplicate language directories from repository root (common/, go/, javascript-typescript/, python/, rust/)
- Fixed nested cli-tool/cli-tool/ directory structure
- All templates now centralized in cli-tool/templates/ (single source of truth)
- Clean repository structure with no duplication
- CLI functionality verified and working correctly
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Updated validation prompt to include project structure analysis
- Claude now examines package.json, src/, and other project files
- Provides project-specific recommendations based on actual dependencies
- Compares installed configuration with real project setup
- Makes more targeted suggestions for improvement
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fixed spawn command to properly execute 'claude "prompt"' format
- Changed from spawn('claude', [prompt]) to spawn('sh', ['-c', claudeCommand])
- Now correctly passes validation prompt to Claude Code CLI
- Post-installation validation now works as intended
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Made validation prompt more explicit about reviewing CLAUDE.md, .claude/settings.json, .claude/commands/, and .mcp.json
- Fixed command structure for proper Claude Code CLI integration
- Improved error handling for validation process
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fixed version command to show correct version (1.3.5)
- Changed bin/create-claude-config.js to read version from package.json
- Version 1.3.4 was published but showed wrong version number
- Now --version will show the actual published version
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Added runPostInstallationValidation function to file-operations.js
- Integrated automatic validation prompt after successful installation
- Claude Code now reviews installed configurations automatically
- Added comprehensive validation prompt with project-specific context
- Users can skip validation or run it to ensure proper setup
- Updated CLI README with validation documentation step 9
- Added validation feature to Features section
- Graceful error handling when Claude Code CLI not available
New onboarding flow:
1. Language selection
2. Framework selection
3. Command selection
4. Hook selection
5. MCP server selection
6. Final confirmation
7. Installation
8. Post-installation validation 🆕🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Added step 6 "MCP Server Selection" with examples of available servers
- Updated final confirmation to include MCP count
- Updated installation output to show MCP server installation
- Added MCP Servers section explaining Model Context Protocol capabilities
- Updated Core Files to include .mcp.json configuration
- Improved development guidelines for open source contributions
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Moved all templates (common, javascript-typescript, python, go, rust) to cli-tool/templates/
- Updated command-scanner.js, file-operations.js, and hook-scanner.js to use local templates
- Added templates/ to package.json files array for npm publishing
- Fixed React and all framework commands not appearing in CLI selection
- All templates now properly bundled with npm package
This resolves the issue where npx claude-code-templates@latest showed empty command selection
because templates were not included in the published package.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fixed React commands not appearing in CLI selection interface
- Removed CustomCheckboxPrompt class that was interfering with choice display
- React commands (component, hooks, state-management) now show correctly
- All interactive steps (commands, hooks, MCPs) now work properly
- Verified with comprehensive testing suite
Resolves issue where command selection appeared empty despite commands being available.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Created a new markdown file for Flask route generation with examples and best practices.
- Added comprehensive testing suite documentation for Flask applications using pytest.
- Included project configuration guidelines for Flask applications.
- Introduced a new MCP configuration file for Rust SDKs and services.
- Updated README.md to reflect new features including command and hook selection.
- Bumped version to 1.2.1 in package.json and updated description.
- Implemented hook scanning functionality in hook-scanner.js to manage automation hooks.
- Modified file-operations.js to handle settings.json with selected hooks.
- Added prompts for hook selection in prompts.js, allowing users to choose automation hooks interactively.
- Enhanced templates for JavaScript, Python, Go, and Rust to include optimized configurations and hooks.
- Updated test scripts to validate hooks functionality and ensure proper configuration in settings.json.
- Improved command structures in settings.json for JavaScript/TypeScript to enhance functionality.
Added complete testing suite for CLI commands including:
- Basic and detailed test scripts with automated scenarios
- Makefile for easy test execution
- NPM scripts integration with pretest hooks
- Testing guide with manual and automated procedures
- Pre-publish test validation in package.json
- Coverage for all language/framework combinations
- Error scenario testing and edge cases
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Dani