Files
Claude-Code-Workflow/.claude/commands/cli/mode/code-analysis.md
catlog22 029f3a3c12 refactor: consolidate CLI commands and templates structure
- Consolidate Gemini, Qwen, and Codex commands into unified CLI commands
- Add new code-analysis mode template
- Update context-gather documentation
- Remove redundant tool-specific command files

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

Co-Authored-By: Claude <noreply@anthropic.com>
2025-09-30 15:52:26 +08:00

7.4 KiB

name, description, usage, argument-hint, examples, allowed-tools, model
name description usage argument-hint examples allowed-tools model
code-analysis Deep code analysis and debugging using CLI tools with specialized template /cli:mode:code-analysis [--tool <codex|gemini|qwen>] [--enhance] [--cd "path"] "analysis target" [--tool codex|gemini|qwen] [--enhance] [--cd path] analysis target
/cli:mode:code-analysis "analyze authentication flow logic"
/cli:mode:code-analysis --tool qwen --enhance "explain data transformation pipeline"
/cli:mode:code-analysis --tool codex --cd "src/core" "trace execution path for user registration"
SlashCommand(*), Bash(*) sonnet

CLI Mode: Code Analysis (/cli:mode:code-analysis)

Purpose

Execute systematic code analysis and debugging using CLI tools with specialized code analysis template.

Supported Tools: codex, gemini (default), qwen

Execution Flow

  1. Parse tool selection: Extract --tool flag (default: gemini)
  2. If --enhance flag present: Execute /enhance-prompt "[analysis-target]" first
  3. Parse analysis target (original or enhanced)
  4. Detect target directory (from --cd or auto-infer)
  5. Build command for selected tool with code-analysis template
  6. Execute deep analysis
  7. Save to session (if active)

Core Rules

  1. Tool Selection: Use --tool value or default to gemini
  2. Enhance First (if flagged): Execute /enhance-prompt before analysis
  3. Directory Context: Use cd when --cd provided or auto-detected
  4. Template Required: Always use code-analysis template
  5. Session Output: Save to .workflow/WFS-[id]/.chat/code-analysis-[timestamp].md

Analysis Capabilities

The code-analysis template provides:

  • Systematic Code Analysis: Break down complex code into manageable parts
  • Execution Path Tracing: Track variable states and call stacks
  • Control & Data Flow: Understand code logic and data transformations
  • Call Flow Visualization: Diagram function calling sequences
  • Logical Reasoning: Explain "why" behind code behavior
  • Debugging Insights: Identify potential bugs or inefficiencies

Command Templates

Core Guidelines: @~/.claude/workflows/intelligent-tools-strategy.md

Gemini (Default)

cd [directory] && ~/.claude/scripts/gemini-wrapper --all-files -p "
PURPOSE: [analysis goal from target]
TASK: Deep code analysis with execution path tracing
CONTEXT: @{CLAUDE.md,**/*CLAUDE.md} [entire codebase in directory]
EXPECTED: Systematic analysis, call flow diagram, data transformations, logical explanation
RULES: $(cat ~/.claude/prompt-templates/code-analysis.md) | Focus on [specific aspect]
"

Qwen

cd [directory] && ~/.claude/scripts/qwen-wrapper --all-files -p "
PURPOSE: [analysis goal from target]
TASK: Architecture-level code analysis and pattern recognition
CONTEXT: @{CLAUDE.md,**/*CLAUDE.md} [entire codebase in directory]
EXPECTED: Architectural insights, design patterns, code structure analysis
RULES: $(cat ~/.claude/prompt-templates/code-analysis.md) | Focus on [specific aspect]
"

Codex

codex -C [directory] --full-auto exec "
PURPOSE: [analysis goal from target]
TASK: Deep code inspection with debugging insights
CONTEXT: @{CLAUDE.md,**/*CLAUDE.md} [entire codebase in directory]
EXPECTED: Execution trace, bug identification, optimization opportunities
RULES: $(cat ~/.claude/prompt-templates/code-analysis.md) | Focus on [specific aspect]
" --skip-git-repo-check -s danger-full-access

Examples

Basic Code Analysis (Gemini):

cd . && ~/.claude/scripts/gemini-wrapper --all-files -p "
PURPOSE: Analyze authentication flow logic
TASK: Trace authentication execution path and identify key functions
CONTEXT: @{CLAUDE.md,**/*CLAUDE.md}
EXPECTED: Step-by-step flow, call diagram, data passing between functions
RULES: $(cat ~/.claude/prompt-templates/code-analysis.md) | Focus on control flow and security
"

Architecture Analysis (Qwen):

# User: /cli:mode:code-analysis --tool qwen "explain data transformation pipeline"

cd . && ~/.claude/scripts/qwen-wrapper --all-files -p "
PURPOSE: Explain data transformation pipeline architecture
TASK: Analyze data flow and transformation patterns
CONTEXT: @{CLAUDE.md,**/*CLAUDE.md}
EXPECTED: Pipeline structure, transformation stages, data format changes
RULES: $(cat ~/.claude/prompt-templates/code-analysis.md) | Focus on data flow and patterns
"

Deep Debugging (Codex):

# User: /cli:mode:code-analysis --tool codex --cd "src/core" "trace execution path for user registration"

codex -C src/core --full-auto exec "
PURPOSE: Trace execution path for user registration
TASK: Deep analysis of registration flow with debugging insights
CONTEXT: @{CLAUDE.md,**/*CLAUDE.md}
EXPECTED: Complete execution trace, variable states, potential issues
RULES: $(cat ~/.claude/prompt-templates/code-analysis.md) | Focus on edge cases and error handling
" --skip-git-repo-check -s danger-full-access

With Enhancement:

# User: /cli:mode:code-analysis --enhance "why is login slow"

# Step 1: Enhance
/enhance-prompt "why is login slow"
# Returns:
# INTENT: Identify performance bottlenecks in login flow
# CONTEXT: Authentication module, database queries
# ACTION: Trace execution path → identify slow operations → suggest optimizations

# Step 2: Analyze with enhanced context
cd . && ~/.claude/scripts/gemini-wrapper --all-files -p "
PURPOSE: Identify performance bottlenecks in login flow
TASK: Trace login execution path and measure operation costs
CONTEXT: @{CLAUDE.md,**/*CLAUDE.md} @{**/*auth*,**/*login*}
EXPECTED: Performance analysis, bottleneck identification, optimization recommendations
RULES: $(cat ~/.claude/prompt-templates/code-analysis.md) | Focus on performance and database queries
"

Analysis Output Structure

Based on code-analysis.md template, output includes:

1. 思考过程 (Thinking Process)

  • Analysis strategy and approach
  • Key assumptions about code behavior

2. 对问题的理解 (Understanding)

  • Restate analysis target
  • Confirm understanding of requirements

3. 核心解答 (Core Answer)

  • Direct, concise answer to analysis question

4. 详细分析与调用逻辑 (Detailed Analysis)

  • 代码段识别: Relevant code sections
  • 执行流程: Step-by-step execution flow
  • 调用图: Visual call flow diagram with symbols:
    • ───► Function call
    • ◄─── Return
    • Continuation
    • ├─ Intermediate step
    • └─ Last step in block
  • 数据传递: Data passing and state changes
  • 逻辑解释: Why code behaves this way

5. 总结 (Summary)

  • Key findings and recommendations

Session Output

Location: .workflow/WFS-[topic]/.chat/code-analysis-[timestamp].md

Includes:

  • Analysis target
  • Template used
  • Complete structured analysis
  • Call flow diagrams
  • Debugging insights
  • Recommendations

Use Cases

Use Case Best Tool Focus
Understand execution flow gemini Call sequences, data flow
Architectural patterns qwen Design patterns, structure
Performance debugging codex Bottlenecks, optimizations
Bug investigation codex Error paths, edge cases
Code review gemini Logic correctness, clarity
Refactoring planning qwen Structure improvements

Tool Selection Guide

  • Gemini: Best for general code understanding and tracing
  • Qwen: Best for architectural analysis and pattern recognition
  • Codex: Best for deep debugging and performance analysis