multi-ai-consultant（多 AI 顾问）

Multi-AI Consultant

Consult external AIs for second opinions when Claude Code is stuck or making critical decisions.

What This Skill Does

This skill enables future Claude Code sessions to consult other AIs when:

• Stuck on a bug after one failed attempt
• Making architectural decisions
• Security concerns need validation
• Fresh perspective needed

Key innovation: Uses existing CLI tools (gemini, codex) instead of building MCP servers - much simpler and more maintainable.

When to Use This Skill

Automatic Triggers (No User Action Needed)

Claude Code should automatically suggest using this skill when:

1. After 1 failed debugging attempt

- Tried one approach to fix a bug - Still not working or different error - → Suggest: "Should I consult [Gemini|Fresh Claude] for a second opinion?"

1. Before architectural decisions

- Significant design choices (state management, routing, data flow) - Framework selection - Database schema design - → Auto-consult (mention to user): "Consulting Gemini for architectural validation..."

1. Security changes

- Authentication logic - Authorization rules - Cryptography - Input validation - → Auto-consult: "Consulting Gemini to verify security approach..."

1. When uncertain

- Multiple valid approaches - Trade-offs not clear - Conflicting advice in documentation - → Suggest: "Would you like me to consult another AI for additional perspective?"

Manual Invocation (User Commands)

User can explicitly request consultation with:

• /consult-gemini [question] - Gemini 2.5 Pro with thinking, search, grounding
• /consult-codex [question] - OpenAI GPT-4 via Codex CLI (repo-aware)
• /consult-claude [question] - Fresh Claude subagent (free, fast)
• /consult-ai [question] - Router that asks which AI to use

The Three AIs

For detailed AI comparison: Load references/ai-strengths.md when choosing which AI to consult for specific use cases.

How It Works

Architecture

Claude Code encounters bug/decision
        ↓
Suggests consultation (or user requests)
        ↓
User approves
        ↓
Execute appropriate slash command
        ↓
CLI command calls external AI
        ↓
Parse response
        ↓
Synthesize: Claude's analysis + External AI's analysis
        ↓
Present 5-part comparison
        ↓
Ask permission to implement

The 5-Part Synthesis Format

Every consultation must follow this format (prevents parroting):

1. 🤖 My Analysis - Claude's original reasoning and attempts
2. 💎/🔷/🔄 Other AI's Analysis - External AI's complete response
3. 🔍 Key Differences - Agreement, divergence, what each AI caught/missed
4. ⚡ Synthesis - Combined perspective, root cause, trade-offs
5. ✅ Recommended Action - Specific next steps with file paths/line numbers

End with: "Should I proceed with this approach?"

Setup

For complete installation guide: Load references/setup-guide.md when installing CLIs, configuring API keys, or setting up templates.

Quick setup:

1. Install CLIs: bun add -g @google/generative-ai-cli (Gemini), bun add -g codex (Codex, optional)
2. Set API keys: export GEMINI_API_KEY="...", export OPENAI_API_KEY="..."
3. Install skill: Symlink to ~/.claude/skills/multi-ai-consultant
4. Copy templates: GEMINI.md, codex.md, .geminiignore to project root
5. Verify: gemini -p "test", codex exec - --yolo

Get API keys:

• Gemini: https://aistudio.google.com/apikey
• OpenAI: https://platform.openai.com/api-keys

Usage

For detailed examples: Load references/usage-examples.md when learning consultation workflows or seeing real-world scenarios.

Quick examples:

• Bug: After 1 failed attempt → /consult-gemini for web-researched solution
• Architecture: Design decision → /consult-gemini for latest best practices
• Code review: Refactoring validation → /consult-codex for repo-aware consistency check
• Quick opinion: Sanity check → /consult-claude for free fresh perspective

5 detailed examples available:

1. JWT authentication bug (saved ~30 min, found platform-specific issue)
2. State management choice (informed decision with 2025 patterns)
3. Refactoring review (found 3 consistency issues)
4. Security validation (found 2 critical issues via OWASP 2025)
5. Multi-AI workflow (high-stakes database choice)

Slash Commands

For complete command reference: Load references/commands-reference.md when needing detailed syntax, options, or cost tracking information.

Quick command overview:

/consult-gemini [question]

• Use: Web research, latest docs, extended thinking
• Features: Google Search, grounding, thinking mode
• Cost: ~$0.10-0.50
• Example: /consult-gemini Is this JWT secure by 2025 standards?

/consult-codex [question]

• Use: Repo-aware analysis, code review
• Features: Auto-scans directory, consistency checks
• Cost: ~$0.05-0.30
• Example: /consult-codex Review for performance bottlenecks

/consult-claude [question]

• Use: Quick second opinion, budget-friendly
• Features: Free, fast, fresh perspective
• Cost: Free
• Example: /consult-claude Am I missing something obvious?

/consult-ai [question]

• Use: Router (recommends which AI to use)
• Features: Analyzes question, suggests best AI
• Cost: Varies by chosen AI
• Example: /consult-ai How should we structure this architecture?

Templates

Templates customize AI behavior for consultations (auto-loaded from project root):

1. GEMINI.md - System instructions for Gemini (enforces 5-part format, web search)
2. codex.md - System instructions for Codex (enforces repo-aware analysis)
3. .geminiignore - Privacy exclusions beyond .gitignore
4. consultation-log-parser.sh - View consultation history (optional)

Installation: Copy from ~/.claude/skills/multi-ai-consultant/templates/ to project root

Privacy & Security

Automatic protection:

• Both CLIs respect .gitignore automatically
• Create .geminiignore for extra exclusions (.env*, *secret*, *credentials*)
• Pre-consultation check warns if sensitive patterns detected

Privacy best practices:

• Always configure .gitignore properly
• Create .geminiignore for extra safety
• Use smart context selection (specific files, not entire repo)
• Verify what will be sent: git status --ignored

For detailed privacy configuration: Load references/setup-guide.md when setting up .geminiignore or privacy exclusions.

Cost Tracking

Every consultation logged to ~/.claude/ai-consultations/consultations.log

Log format: timestamp,ai,model,input_tokens,output_tokens,cost,project_path

View logs: consultation-log-parser.sh --summary

Example output:

Total consultations: 47
Gemini: 23 ($4.25), Codex: 12 ($1.85), Fresh Claude: 12 ($0.00)
Total cost: $6.10

For detailed cost tracking: Load references/commands-reference.md when viewing logs, calculating costs, or managing budgets.

Common Issues

For complete troubleshooting: Load references/troubleshooting.md when encountering errors or setup issues.

Top 5 issues:

1. CLI not installed: gemini: command not found → Fix: bun add -g @google/generative-ai-cli
2. API key invalid: Authentication failed → Fix: export GEMINI_API_KEY="..."
3. Context too large: Token limit exceeded → Fix: Use specific files, not entire repo
4. Privacy leak: .env file sent → Fix: Add to .gitignore and .geminiignore
5. Skill not discovered: Not working → Fix: Check ~/.claude/skills/multi-ai-consultant

Token Efficiency

Without This Skill

Typical scenario (stuck on bug):

1. Try approach 1 (~4k tokens)
2. Research CLI syntax (~3k tokens)
3. Try approach 2 (~4k tokens)
4. Research documentation (~3k tokens)
5. Try approach 3 (~4k tokens)

Total: ~20k tokens, 30-45 minutes

With This Skill

Same scenario:

1. Try approach 1 (~4k tokens)
2. Execute /consult-gemini (~1k tokens)
3. Gemini finds issue (<5k tokens, billed separately)
4. Implement fix (~3k tokens)

Total: ~8k tokens, 5-10 minutes

Savings: ~60% tokens, ~75% time

Success Metrics

Time Efficiency

• Without skill: 30-45 minutes (trial and error)
• With skill: 5-10 minutes (consultation + fix)
• Savings: ~75%

Token Efficiency

• Without skill: ~20k tokens (multiple attempts)
• With skill: ~8k tokens (one consultation)
• Savings: ~60%

Error Prevention

• Manual CLI use: 3-5 common errors (flags, parsing, privacy)
• With skill: 0 errors (all handled by commands)
• Prevention: 100%

Quality

• Manual: Risk of not synthesizing (just copying external AI)
• With skill: Forced synthesis via GEMINI.md/codex.md
• Improvement: Guaranteed value-add

Why CLI Approach (Not MCP)?

Winner: CLI approach - 80% less effort, same functionality

When to Load References

Load reference files when working on specific aspects of AI consultation:

ai-strengths.md

Load when:

• Selection-based: Choosing which AI to consult (Gemini vs Codex vs Fresh Claude)
• Comparison-based: Understanding capabilities, costs, and trade-offs between AIs
• Strategy-based: Planning combination strategies (free → paid, paid first, budget-conscious)
• Capability-based: Understanding special features (Google Search, extended thinking, grounding, repo-aware, fresh perspective)
• Scenario-based: Multiple AI consultation workflows, validation workflows

setup-guide.md

Load when:

• Installation-based: Setting up Gemini CLI, Codex CLI, or installing the skill
• Configuration-based: Configuring API keys, environment variables, or system paths
• Privacy-based: Setting up.geminiignore, privacy exclusions, or security configurations
• Template-based: Installing GEMINI.md, codex.md,.geminiignore, or consultation-log-parser.sh
• Verification-based: Testing CLI installation, API keys, or skill discovery

commands-reference.md

Load when:

• Command-based: Using /consult-gemini, /consult-codex, /consult-claude, or /consult-ai
• Syntax-based: Understanding command flags, options, or context selection
• Cost-based: Understanding cost tracking, log format, or viewing consultation history
• Logging-based: Using consultation-log-parser.sh or analyzing consultation patterns

usage-examples.md

Load when:

• Scenario-based: Learning how to use skill for bugs, architecture decisions, or code review
• Workflow-based: Understanding consultation workflow, synthesis process, or multi-AI approach
• Example-based: Seeing real-world examples of consultations and their outcomes (5 detailed examples available)

troubleshooting.md

Load when:

• Error-based: Encountering specific errors (command not found, API key invalid, parsing failures)
• Diagnosis-based: Troubleshooting CLI issues, API connectivity, or skill discovery
• Fix-based: Resolving known issues with step-by-step solutions (8 common issues documented)
• Debugging-based: Testing CLIs manually, checking configurations, or verifying installations

Contributing

Found an issue?

• Document it in troubleshooting.md
• Include fix/workaround
• Update slash commands to prevent

Adding new AI?

• Create new slash command: commands/consult-newai.md
• Add to router: Update commands/consult-ai.md
• Create template: templates/newai.md (if CLI supports system instructions)
• Update documentation

Improving synthesis?

• Edit templates: templates/GEMINI.md, templates/codex.md
• Test with real consultations
• Measure before/after quality

References

External Resources

• Gemini CLI: https://ai.google.dev/gemini-api/docs/cli
• OpenAI Codex: https://www.npmjs.com/package/codex
• OpenAI API: https://platform.openai.com/docs
• Gemini API Pricing: https://ai.google.dev/pricing
• OpenAI Pricing: https://openai.com/pricing

Internal Files

• Planning docs: planning/multi-ai-consultant-*.md
• Slash commands: commands/*.md
• Templates: templates/*
• Scripts: scripts/*
• References: references/*.md (5 reference files)

License

MIT License - See LICENSE file

Last Updated: 2025-11-07 Status: Production Ready Maintainer: Claude Skills Maintainers | maintainers@example.com