- Deleted advanced workflows, analysis examples, context optimization, template usage, and write examples for both Gemini and Qwen to streamline documentation.
- Updated CLAUDE.md to include references to new context strategies and search commands.
- Enhanced clarity and focus on compliance, security, and performance in remaining documentation.
- Gemini: add Chinese task examples (分析代码库, 理解架构, etc.)
- Qwen: add Chinese task labels (分析, 探索, 文档任务, etc.)
- Codex: add Chinese task examples (实现功能, 修复bug, 重构代码, etc.)
- Enable auto-trigger for both English and Chinese keywords
Add comprehensive skill system and batch replan functionality:
- New skill modules for Gemini, Qwen, Codex, and context search
- Enhanced task replan command with batch mode
- Improved action plan verification
- Complete skill documentation with examples
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Enhanced layout-extract and style-extract workflows with two-phase generation
process, allowing users to preview and select preferred options before full
generation.
Key changes:
- Split generation into two agent tasks:
* Task 1: Generate concept options with previews
* Task 2: Develop selected option into complete output
- Add Phase 1.5 interactive user selection:
* Present concept options with visual previews
* Capture user selection via AskUserQuestion tool
* Save selection to user-selection.json
- Layout workflow improvements:
* Generate layout concept options with ASCII wireframes
* User selects preferred structural approach
* Generate detailed templates only for selected concepts
- Style workflow improvements:
* Generate design direction options with color/typography previews
* User selects preferred design philosophy
* Generate complete design system only for selected direction
- Better user experience:
* Avoid generating unwanted variants
* Allow informed decisions before full generation
* Reduce wasted computation on unused variants
This change improves workflow efficiency and user control over design output.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Renamed test-gen to test-fix-gen and created new test-execute command for
dynamic iterative test-fix cycles. Improved task generation flexibility and
clarified data flow between planning and execution phases.
Key changes:
- Rename test-gen.md → test-fix-gen.md to reflect task generation purpose
- Add test-execute.md for dynamic test-fix workflow execution with CLI analysis
- Enhance test-fix-gen.md:
* Emphasize code-developer's understanding and analysis responsibilities
* Support flexible task count (minimum 2, expandable based on complexity)
* Add examples for multi-module test generation scenarios
- Clarify test-execute.md data flow:
* failure_context contains raw test output and error messages
* fix_strategy generated by CLI tool (Gemini/Qwen) analysis
* IMPL-001 (test generation) executes first, then IMPL-002 (test-fix cycle)
This improves workflow clarity and enables adaptive test-fix strategies for
complex projects requiring multiple test generation tasks.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Fixes critical P0 issue where animation-tokens.json wasn't consumed by the
generate command, breaking the value chain. The animation extraction system
now properly flows through: animation-extract → tokens → generate → prototypes.
Changes:
- Added animation-extract command with hybrid CSS extraction + interactive
fallback strategy
- Updated generate.md to load and inject animation tokens into prototypes
- Added CSS animation support (custom properties, keyframes, interactions,
accessibility)
- Integrated animation extraction into explore-auto and imitate-auto workflows
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Automatically updated by linter to reflect new --count parameter
in argument hint for command usage clarity.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
**Problem**: synthesis.md incorrectly used action-planning-agent
- synthesis-specification.md is conceptual analysis document (WHAT to build)
- action-planning-agent is designed for execution plans (HOW to build)
- Mismatch between agent capabilities and task requirements
**Solution**: Change to conceptual-planning-agent
- Aligns with task nature (conceptual synthesis and multi-perspective integration)
- Consistent with other role analysis commands (data-architect, system-architect, etc.)
- Agent specializes in conceptual analysis rather than task decomposition
**Changes**:
- Update allowed-tools: action-planning-agent → conceptual-planning-agent
- Update description to note agent usage
- Add "Conceptual Focus" rationale explaining agent specialization
- Update all agent references throughout document (8 locations)
- Maintain consistency with role analysis command pattern
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add --count N parameter to specify number of roles (default: 3, max: 9)
- Add Parameter Parsing section with count handling logic
- Update role selection logic to use dynamic count value
- Update Usage section with parameter documentation
- Update TodoWrite examples to reflect dynamic role count
- Update parallel agent invocation examples with N roles
- Add role ranking and diversity selection guidelines
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add documentation for Chrome DevTools MCP server requirement across
English and Chinese README files:
Changes:
- Add Chrome DevTools MCP to MCP servers table with installation link
- Mark as required for UI workflows (URL mode design extraction)
- Add UI Workflows benefit to MCP benefits section
- Add warning note before UI Design Workflow Commands section
- Document graceful fallback to visual analysis when unavailable
Reference: https://github.com/ChromeDevTools/chrome-devtools-mcp🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add --urls parameter support to style-extract and layout-extract
- Auto-trigger computed styles extraction when URLs provided (no manual flag)
- Auto-trigger DOM structure extraction for layout analysis
- Streamline imitate-auto workflow by removing redundant REPORT statements
- Maintain clear phase headers while reducing verbose output
- Enable graceful fallback to visual analysis if Chrome DevTools unavailable
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Update version badges from v4.4.0 to v4.5.0 in README files
- Add v4.5.0 release highlights:
* Enhanced agent protocols with unified execution
* Streamlined UI workflows and improved documentation
* Optimized workflow templates for better clarity
* Improved CLI commands and tool integration
* New CLI execution agent for autonomous task handling
- Update Getting Started guides to reference v4.5.0
- Add cli-execution-agent to agent list in Getting Started guides
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Clarified path format requirements:
- MCP tools require double backslash format (D:\path)
- Bash commands support forward slash (D:/path) or POSIX (/d/path)
- Added quick reference for common conversions
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Revert intelligent-tools-strategy.md to the version from commit b5d6870
(docs: enhance method parameter documentation in unified template).
This reverts the following commits for this file:
- 4524702 docs: restore Template System directory tree structure
- a7a157d docs: restore critical details to intelligent-tools-strategy.md
- 1e79866 refactor: optimize intelligent-tools-strategy.md and fix codex calls
File changes: 451 lines → 503 lines (+52 lines net)
- Restores original detailed structure
- Reverts optimization changes
- Returns to pre-refactor state
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Restore the original directory tree visualization for Template System section,
replacing the compressed table format with a clear hierarchical structure.
Changes:
- Replace table format with ASCII directory tree
- Show complete file paths: prompts/, planning-roles/, tech-stacks/
- Display full template hierarchy with descriptions
- Improve path clarity for RULES field references
Benefits:
- Clear visualization of template file locations
- Easy to reference in $(cat ...) commands
- Better understanding of directory structure
- Facilitates accurate template path construction
File: intelligent-tools-strategy.md: 431 → 451 lines (+20 lines)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Restore important detailed content that was removed in commit 1e79866:
Key restorations (106 lines added):
- Session Management: Detailed explanation and multi-task pattern examples (15 lines)
- Auto-Resume Decision Rules: Complete criteria for when to use/not use (12 lines)
- RULES Field Format: Critical "template reference only, never read" rule (14 lines)
- File Pattern Reference: Complete 3-step workflow with rg/MCP examples (33 lines)
- Common Scenarios: Full examples including Feature Development 3-step workflow (82 lines)
- Best Practices: Detailed guidelines with context optimization strategy (8 lines)
Additional improvements:
- Enhanced CLAUDE.md references throughout the document
- Added CLAUDE.md to Planning Checklist and File Pattern examples
- Restored execution timeout configuration with specific millisecond values
File changes:
- intelligent-tools-strategy.md: 325 → 431 lines (balanced detail/conciseness)
- Removed design-tokens-schema.md (obsolete)
The document now maintains conciseness while restoring all critical guidance.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Reduce intelligent-tools-strategy.md from 504 to 325 lines (35.5% reduction)
- Merge Gemini/Qwen tool specifications into unified Analysis Tools section
- Consolidate command templates and remove duplicate examples
- Compress usage scenarios while preserving all critical information
- Simplify guidelines to concise bullet points
- Fix incorrect codex chat calls in discuss-plan.md to use proper exec format
All core decision rules, permission framework, and command templates preserved.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add comprehensive requirements for documenting method parameters, return values, and exceptions in code files. Include detailed guidelines for parameter types, descriptions, default values, and usage examples.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Removed the "Direct Gemini Wrapper Script Usage" section from Getting Started guides
as users cannot manually execute these commands in Claude Code - all tool invocations
are handled automatically by Claude through semantic understanding.
Changes:
- Removed "Gemini Wrapper 脚本直接使用" section (Chinese)
- Removed "Direct Gemini Wrapper Script Usage" section (English)
- Kept "Semantic Tool Invocation" section explaining natural language usage
- Streamlined documentation to focus on user-facing interaction patterns
Rationale:
- Users cannot run ~/.claude/scripts/gemini-wrapper commands directly in Claude Code
- These are internal implementation details handled automatically by Claude
- Documentation should focus on how users interact with the system, not internal mechanics
Files updated:
- GETTING_STARTED.md (-47 lines)
- GETTING_STARTED_CN.md (-47 lines)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Fixed the "Semantic Tool Invocation" concept to accurately reflect its meaning:
**Before**: "Gemini 工具语义调用" referred to direct script usage
**After**: Properly describes semantic invocation as natural language interaction
Changes:
- Added "Semantic Tool Invocation" section explaining natural language usage
- Example: User says "Use gemini to analyze architecture"
- Claude understands and automatically executes appropriate commands
- Highlighted advantages: natural interaction, intelligent understanding, auto optimization
- Renamed previous section to "Direct Gemini Wrapper Script Usage"
- Clarified this is for scenarios requiring precise control
- Maintained all existing script usage examples
Structure now clearly distinguishes:
1. Semantic invocation (natural language → Claude → tool execution)
2. Direct script usage (manual command execution)
Updated both:
- GETTING_STARTED.md (English)
- GETTING_STARTED_CN.md (Chinese)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Added eye-catching links to the beginner-friendly Getting Started guides
at the top of both README files, right after the version announcement.
This helps new users quickly discover the 5-minute tutorial.
Changes:
- README.md: Added link to GETTING_STARTED.md
- README_CN.md: Added link to GETTING_STARTED_CN.md
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Updated README documentation to emphasize that index.html and compare.html
can be opened directly in browser without requiring a local server.
Reorganized preview instructions to highlight direct browser preview as
the recommended approach, with local server as optional for advanced features.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Convert all .sh files from CRLF to LF line endings
- Add .gitattributes to enforce LF for shell scripts
- Fix "bash\r: No such file or directory" error in WSL environments
This resolves the installation failure on WSL/Linux systems where
Windows CRLF line endings cause shell scripts to fail.
Related to #1🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add semantic file discovery via Gemini CLI --all-files parameter
- Define codebase-retrieval as primary semantic search tool
- Add tool selection matrix for search strategy
- Provide command templates with dynamic placeholders
- Document workflow integration with rg and other tools
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Update enhance-prompt.md to use correct file reference format
- Change from @~/.claude/workflows/intelligent-tools-strategy.md to ~/.claude/workflows/intelligent-tools-strategy.md
- Align with other command files' reference format
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Replace 4-layer hierarchy (Layer 1-4) with single unified template
- New template: claude-module-unified.txt works for all modules and files
- Update update_module_claude.sh to remove layer detection logic
- Archive old layer-based templates to memory/archive/
- Update intelligent-tools-strategy.md to reference unified template
Benefits:
✅ Single template to maintain
✅ Simpler script logic
✅ Universal application for folders and code files
✅ Clearer documentation structure
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Key changes:
1. Add CRITICAL section for command substitution rules
- Document why escape characters (\$, \", \') break commands
- Explain correct syntax without quotes in $(cat ...) paths
- Provide clear correct vs wrong examples
2. Fix all template references in examples
- Remove single quotes from $(cat '...') → $(cat ...)
- Ensure all 4 occurrences use correct syntax (lines 331, 356, 366, 379)
3. Add best practice guideline
- Add warning in General Guidelines about escape characters
- Emphasize this applies to all CLI command execution
Problem solved: Prevent automatic escape character insertion that breaks
shell command substitution and path expansion in actual CLI execution.
