Files
Claude-Code-Workflow/.claude/skills/command-guide/SKILL.md
catlog22 38b070551c fix: 使用绝对路径确保 skill 在全局安装时正常工作
问题
- 原有代码使用相对路径 (cd reference &&)
- skill 安装在 ~/.claude/skills/ 后相对路径会失效

修复
- 所有文件操作使用绝对路径: ~/.claude/skills/command-guide/reference/
- CLI 命令改用 --include-directories 参数指向绝对路径
- 更新示例输出中的路径为绝对路径

影响
- handleSimpleQuery: basePath 使用绝对路径
- locateCommandFile: 使用 basePath 参数
- executeCLIAnalysis: 使用 --include-directories 替代 cd
- resolveEntityPath: glob 使用绝对路径
- buildCLIPrompt: CONTEXT 使用 @**/* + --include-directories

版本更新
- v1.3.0 → v1.3.1

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-06 16:37:21 +08:00

10 KiB

name, description, allowed-tools
name description allowed-tools
command-guide Workflow command guide for Claude DMS3 (69 commands). Search/browse commands, get next-step recommendations, view documentation, report issues. Triggers "CCW-help", "CCW-issue", "ccw-help", "ccw-issue", "ccw" Read, Grep, Glob, AskUserQuestion

Command Guide Skill

Comprehensive command guide for Claude DMS3 workflow system covering 69 commands across 4 categories (workflow, cli, memory, task).

🎯 Operation Modes

Mode 1: Command Search 🔍

When: User searches by keyword, category, or use-case

Triggers: "搜索命令", "find command", "planning 相关", "search"

Process:

  1. Identify search type (keyword/category/use-case)
  2. Query appropriate index (all-commands/by-category/by-use-case)
  3. Return matching commands with metadata
  4. Suggest related commands

Example: "搜索 planning 命令" → Lists planning commands from index/by-use-case.json


Mode 2: Smart Recommendations 🤖

When: User asks for next steps after a command

Triggers: "下一步", "what's next", "after /workflow:plan", "推荐"

Process:

  1. Parse current context (last command/workflow state)
  2. Query index/command-relationships.json
  3. Return recommended next commands with rationale
  4. Show common workflow patterns

Example: "执行完 /workflow:plan 后做什么?" → Recommends /workflow:execute or /workflow:action-plan-verify


Mode 3: Full Documentation 📖

When: User requests command details

Triggers: "参数说明", "怎么用", "how to use", "详情"

Process:

  1. Locate command in index/all-commands.json
  2. Read original command file for full details
  3. Present parameters, arguments, examples
  4. Link to related commands

Example: "/workflow:plan 的参数是什么?" → Shows full parameter list and usage examples


Mode 4: Beginner Onboarding 🎓

When: New user needs guidance

Triggers: "新手", "getting started", "如何开始", "常用命令"

Process:

  1. Present progressive learning path
  2. Show index/essential-commands.json (Top 14 commands)
  3. Link to getting-started guide
  4. Provide first workflow example

Example: "我是新手,如何开始?" → Learning path + Top 14 commands + quick start guide


Mode 5: Issue Reporting 📝

When: User wants to report issue or request feature

Triggers: "CCW-issue", "CCW-help", "ccw-issue", "ccw-help", "ccw", "报告 bug", "功能建议", "问题咨询", "交互式诊断"

Process:

  1. Use AskUserQuestion to confirm type (diagnosis/bug/feature/question)
  2. Collect required information interactively with execution flow emphasis
  3. Select appropriate template:
    • issue-diagnosis.md - Full diagnostic workflow with decision tree
    • issue-bug.md - Bug report with complete command history
    • issue-feature.md - Feature request with current workflow analysis
    • issue-question.md - Question with detailed attempt history
  4. Generate filled template with privacy-protected command history
  5. Save/display with troubleshooting guidance

Example: "CCW-issue" → Interactive Q&A → Generates GitHub issue template with full execution context

🆕 Enhanced Features:

  • Complete command history with privacy protection
  • Interactive diagnostic checklists
  • Decision tree navigation (diagnosis template)
  • Execution environment capture

Mode 6: Deep Command Analysis 🔬

When: User asks detailed questions about specific commands or agents

Triggers: "详细说明", "命令原理", "agent 如何工作", "实现细节", specific command/agent name mentioned

Data Sources:

  • reference/agents/*.md - All agent documentation (11 agents)
  • reference/commands/**/*.md - All command documentation (69 commands)

Process:

Simple Query (direct documentation lookup):

  1. Identify target command/agent from user query
  2. Locate corresponding markdown file in reference/
  3. Read and extract relevant sections
  4. Present formatted response with examples

Complex Query (CLI-assisted analysis):

  1. Detect complexity indicators (多个命令对比、工作流程分析、最佳实践)
  2. Construct analysis prompt for gemini/qwen:
    gemini -p "
    PURPOSE: Analyze command documentation to answer user query
    TASK: [extracted user question with context]
    MODE: analysis
    CONTEXT: @**/*
    EXPECTED: Comprehensive answer with examples and recommendations
    RULES: $(cat ~/.claude/workflows/cli-templates/prompts/analysis/02-analyze-code-patterns.txt) | Focus on practical usage | analysis=READ-ONLY
    " -m gemini-3-pro-preview-11-2025 --include-directories ~/.claude/skills/command-guide/reference
    
    Note: Use absolute path ~/.claude/skills/command-guide/reference for reference documentation access
  3. Return CLI analysis results to user

Query Classification:

  • Simple: Single command explanation, parameter list, basic usage
  • Complex: Cross-command workflows, performance comparison, architectural analysis, best practices across multiple commands

Examples:

Simple Query:

User: "action-planning-agent 如何工作?"
→ Read reference/agents/action-planning-agent.md
→ Extract workflow, capabilities, examples
→ Present structured response

Complex Query:

User: "对比 workflow:plan 和 workflow:tdd-plan 的使用场景和最佳实践"
→ Detect: 多命令对比 + 最佳实践
→ Use gemini analysis on reference/commands/workflow/
→ Return comprehensive comparison with workflow examples

📚 Index Files

All command metadata is stored in JSON indexes for fast querying:

  • all-commands.json - Complete catalog (69 commands) with full metadata
  • by-category.json - Hierarchical organization (workflow/cli/memory/task)
  • by-use-case.json - Grouped by scenario (planning/implementation/testing/docs/session)
  • essential-commands.json - Top 14 most-used commands
  • command-relationships.json - Next-step recommendations and dependencies

📖 Detailed structure: Index Structure Reference


🗂️ Supporting Guides

📦 Reference Documentation

Complete backup of all command and agent documentation for deep analysis:

  • reference/agents/ - 11 agent markdown files with implementation details
  • reference/commands/ - 69 command markdown files organized by category
    • cli/ - CLI tool commands (9 files)
    • memory/ - Memory management commands (8 files)
    • task/ - Task management commands (4 files)
    • workflow/ - Workflow commands (46 files)

Installation Path: ~/.claude/skills/command-guide/ (skill designed for global installation)

Absolute Reference Path: ~/.claude/skills/command-guide/reference/

Usage: Mode 6 queries these files directly for detailed command/agent analysis, or uses CLI tools (gemini/qwen) with absolute paths for complex cross-command analysis.


🛠️ Issue Templates

Generate standardized GitHub issue templates with execution flow emphasis:

  • Interactive Diagnosis - 🆕 Comprehensive diagnostic workflow with decision tree, checklists, and full command history
  • Bug Report - Report command errors with complete execution flow and environment details
  • Feature Request - Suggest improvements with current workflow analysis and pain points
  • Question - Ask usage questions with detailed attempt history and context

All templates now include:

  • Complete command history sections (with privacy protection)
  • Execution environment details
  • Interactive problem-locating checklists
  • Structured troubleshooting guidance

Templates are auto-populated during Mode 5 (Issue Reporting) interaction.


📊 System Statistics

  • Total Commands: 69
  • Total Agents: 11
  • Categories: 4 (workflow: 46, cli: 9, memory: 8, task: 4, general: 2)
  • Use Cases: 5 (planning, implementation, testing, documentation, session-management)
  • Difficulty Levels: 3 (Beginner, Intermediate, Advanced)
  • Essential Commands: 14
  • Reference Docs: 80 markdown files (11 agents + 69 commands)

🔧 Maintenance

Updating Indexes

When commands are added/modified/removed:

bash scripts/update-index.sh

This script:

  1. Scans all command files in ../../commands/
  2. Extracts metadata from YAML frontmatter
  3. Analyzes command relationships
  4. Regenerates all 5 index files

Committing Updates

git add .claude/skills/command-guide/index/
git commit -m "docs: update command indexes"
git push

Team members get latest indexes via git pull.



Version: 1.3.1 (Path configuration for global installation) Last Updated: 2025-11-06 Maintainer: Claude DMS3 Team

Changelog v1.3.1:

  • Updated all paths to use absolute paths (~/.claude/skills/command-guide/)
  • CLI commands now use --include-directories with absolute reference path
  • Ensures skill works correctly when installed in ~/.claude/skills/

Changelog v1.3.0:

  • Added Mode 6: Deep Command Analysis with CLI-assisted queries
  • Created reference documentation backup (80 files: 11 agents + 69 commands)
  • Support simple queries (direct file lookup) and complex queries (CLI analysis)
  • Integrated gemini/qwen for cross-command analysis and best practices

Changelog v1.2.0:

  • Added Interactive Diagnosis template with decision tree
  • Enhanced all templates with complete command history sections
  • Added privacy protection guidelines for sensitive information
  • Integrated execution flow emphasis across all issue templates