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:
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?"
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..."
Security changes
- Authentication logic
- Authorization rules
- Cryptography
- Input validation
- → Auto-consult: "Consulting Gemini to verify security approach..."
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
| AI |
Tool |
When to Use |
Special Features |
Cost |
| Gemini 2.5 Pro |
gemini CLI |
Web research, latest docs, thinking |
Google Search, extended reasoning, grounding |
~$0.10-0.50 |
| OpenAI GPT-4 |
codex CLI |
Repo-aware analysis, code review |
Auto-scans directory, OpenAI reasoning |
~$0.05-0.30 |
| Fresh Claude |
Task tool |
Quick second opinion, budget-friendly |
Same capabilities, fresh perspective |
Free |
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):
- 🤖 My Analysis - Claude's original reasoning and attempts
- 💎/🔷/🔄 Other AI's Analysis - External AI's complete response
- 🔍 Key Differences - Agreement, divergence, what each AI caught/missed
- ⚡ Synthesis - Combined perspective, root cause, trade-offs
- ✅ 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:
- Install CLIs:
bun add -g @google/generative-ai-cli (Gemini), bun add -g codex (Codex, optional)
- Set API keys:
export GEMINI_API_KEY="...", export OPENAI_API_KEY="..."
- Install skill: Symlink to
~/.claude/skills/multi-ai-consultant
- Copy templates:
GEMINI.md, codex.md, .geminiignore to project root
- Verify:
gemini -p "test", codex exec - --yolo
Get 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:
- JWT authentication bug (saved ~30 min, found platform-specific issue)
- State management choice (informed decision with 2025 patterns)
- Refactoring review (found 3 consistency issues)
- Security validation (found 2 critical issues via OWASP 2025)
- 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):
- GEMINI.md - System instructions for Gemini (enforces 5-part format, web search)
- codex.md - System instructions for Codex (enforces repo-aware analysis)
- .geminiignore - Privacy exclusions beyond
.gitignore
- 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:
- CLI not installed:
gemini: command not found → Fix: bun add -g @google/generative-ai-cli
- API key invalid: Authentication failed → Fix:
export GEMINI_API_KEY="..."
- Context too large: Token limit exceeded → Fix: Use specific files, not entire repo
- Privacy leak:
.env file sent → Fix: Add to .gitignore and .geminiignore
- Skill not discovered: Not working → Fix: Check
~/.claude/skills/multi-ai-consultant
Token Efficiency
Without This Skill
Typical scenario (stuck on bug):
- Try approach 1 (~4k tokens)
- Research CLI syntax (~3k tokens)
- Try approach 2 (~4k tokens)
- Research documentation (~3k tokens)
- Try approach 3 (~4k tokens)
Total: ~20k tokens, 30-45 minutes
With This Skill
Same scenario:
- Try approach 1 (~4k tokens)
- Execute
/consult-gemini (~1k tokens)
- Gemini finds issue (<5k tokens, billed separately)
- 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)?
| Aspect |
MCP Server |
CLI Approach |
| Setup time |
4-6 hours |
60-75 minutes |
| Complexity |
High (MCP protocol) |
Low (bash + CLIs) |
| Maintenance |
Update MCP SDK |
Update CLI (rare) |
| Flexibility |
Locked to AIs |
Any AI with CLI |
| Debugging |
MCP protocol |
Standard bash |
| Dependencies |
MCP SDK, npm |
Just CLIs |
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
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
1---2name: multi-ai-consultant3description: Consult external AIs (Gemini 2.5 Pro, OpenAI Codex, Claude) for second opinions. Use for debugging failures, architectural decisions, security validation, or need fresh perspective with synthesis.4license: MIT5allowed-tools: [Bash, Read, Task, Write]6metadata:7 source: https://skillmd.com/skills/secondsky/multi-ai-consultant8 version: 1.0.09 author: Claude Skills Maintainers10 last-verified: 2025-11-0711 production-tested: true12 keywords:13 - ai consultation14 - second opinion15 - gemini16 - openai17 - codex18 - debugging19 - architecture20 - multi-ai21 - synthesis22 - fresh perspective23 - stuck24 - web research25 - thinking mode26 - repo-aware27---2829# Multi-AI Consultant3031Consult external AIs for second opinions when Claude Code is stuck or making critical decisions.3233---3435## What This Skill Does3637This skill enables **future Claude Code sessions** to consult other AIs when:38- Stuck on a bug after one failed attempt39- Making architectural decisions40- Security concerns need validation41- Fresh perspective needed4243**Key innovation**: Uses existing CLI tools (`gemini`, `codex`) instead of building MCP servers - much simpler and more maintainable.4445---4647## When to Use This Skill4849### Automatic Triggers (No User Action Needed)5051Claude Code should **automatically suggest** using this skill when:52531. **After 1 failed debugging attempt**54 - Tried one approach to fix a bug55 - Still not working or different error56 - → Suggest: "Should I consult [Gemini|Fresh Claude] for a second opinion?"57582. **Before architectural decisions**59 - Significant design choices (state management, routing, data flow)60 - Framework selection61 - Database schema design62 - → Auto-consult (mention to user): "Consulting Gemini for architectural validation..."63643. **Security changes**65 - Authentication logic66 - Authorization rules67 - Cryptography68 - Input validation69 - → Auto-consult: "Consulting Gemini to verify security approach..."70714. **When uncertain**72 - Multiple valid approaches73 - Trade-offs not clear74 - Conflicting advice in documentation75 - → Suggest: "Would you like me to consult another AI for additional perspective?"7677### Manual Invocation (User Commands)7879User can explicitly request consultation with:80- `/consult-gemini [question]` - Gemini 2.5 Pro with thinking, search, grounding81- `/consult-codex [question]` - OpenAI GPT-4 via Codex CLI (repo-aware)82- `/consult-claude [question]` - Fresh Claude subagent (free, fast)83- `/consult-ai [question]` - Router that asks which AI to use8485---8687## The Three AIs8889| AI | Tool | When to Use | Special Features | Cost |90|----|------|-------------|------------------|------|91| **Gemini 2.5 Pro** | `gemini` CLI | Web research, latest docs, thinking | Google Search, extended reasoning, grounding | ~$0.10-0.50 |92| **OpenAI GPT-4** | `codex` CLI | Repo-aware analysis, code review | Auto-scans directory, OpenAI reasoning | ~$0.05-0.30 |93| **Fresh Claude** | Task tool | Quick second opinion, budget-friendly | Same capabilities, fresh perspective | Free |9495**For detailed AI comparison**: Load `references/ai-strengths.md` when choosing which AI to consult for specific use cases.9697---9899## How It Works100101### Architecture102103```104Claude Code encounters bug/decision105 ↓106Suggests consultation (or user requests)107 ↓108User approves109 ↓110Execute appropriate slash command111 ↓112CLI command calls external AI113 ↓114Parse response115 ↓116Synthesize: Claude's analysis + External AI's analysis117 ↓118Present 5-part comparison119 ↓120Ask permission to implement121```122123### The 5-Part Synthesis Format124125Every consultation must follow this format (prevents parroting):1261271. **🤖 My Analysis** - Claude's original reasoning and attempts1282. **💎/🔷/🔄 Other AI's Analysis** - External AI's complete response1293. **🔍 Key Differences** - Agreement, divergence, what each AI caught/missed1304. **⚡ Synthesis** - Combined perspective, root cause, trade-offs1315. **✅ Recommended Action** - Specific next steps with file paths/line numbers132133**End with**: "Should I proceed with this approach?"134135---136137## Setup138139**For complete installation guide**: Load `references/setup-guide.md` when installing CLIs, configuring API keys, or setting up templates.140141**Quick setup**:1421. **Install CLIs**: `bun add -g @google/generative-ai-cli` (Gemini), `bun add -g codex` (Codex, optional)1432. **Set API keys**: `export GEMINI_API_KEY="..."`, `export OPENAI_API_KEY="..."`1443. **Install skill**: Symlink to `~/.claude/skills/multi-ai-consultant`1454. **Copy templates**: `GEMINI.md`, `codex.md`, `.geminiignore` to project root1465. **Verify**: `gemini -p "test"`, `codex exec - --yolo`147148Get API keys:149- Gemini: https://aistudio.google.com/apikey150- OpenAI: https://platform.openai.com/api-keys151152---153154## Usage155156**For detailed examples**: Load `references/usage-examples.md` when learning consultation workflows or seeing real-world scenarios.157158**Quick examples**:159- **Bug**: After 1 failed attempt → `/consult-gemini` for web-researched solution160- **Architecture**: Design decision → `/consult-gemini` for latest best practices161- **Code review**: Refactoring validation → `/consult-codex` for repo-aware consistency check162- **Quick opinion**: Sanity check → `/consult-claude` for free fresh perspective163164**5 detailed examples available**:1651. JWT authentication bug (saved ~30 min, found platform-specific issue)1662. State management choice (informed decision with 2025 patterns)1673. Refactoring review (found 3 consistency issues)1684. Security validation (found 2 critical issues via OWASP 2025)1695. Multi-AI workflow (high-stakes database choice)170171---172173## Slash Commands174175**For complete command reference**: Load `references/commands-reference.md` when needing detailed syntax, options, or cost tracking information.176177**Quick command overview**:178179### /consult-gemini [question]180- **Use**: Web research, latest docs, extended thinking181- **Features**: Google Search, grounding, thinking mode182- **Cost**: ~$0.10-0.50183- **Example**: `/consult-gemini Is this JWT secure by 2025 standards?`184185### /consult-codex [question]186- **Use**: Repo-aware analysis, code review187- **Features**: Auto-scans directory, consistency checks188- **Cost**: ~$0.05-0.30189- **Example**: `/consult-codex Review for performance bottlenecks`190191### /consult-claude [question]192- **Use**: Quick second opinion, budget-friendly193- **Features**: Free, fast, fresh perspective194- **Cost**: Free195- **Example**: `/consult-claude Am I missing something obvious?`196197### /consult-ai [question]198- **Use**: Router (recommends which AI to use)199- **Features**: Analyzes question, suggests best AI200- **Cost**: Varies by chosen AI201- **Example**: `/consult-ai How should we structure this architecture?`202203---204205## Templates206207Templates customize AI behavior for consultations (auto-loaded from project root):2082091. **GEMINI.md** - System instructions for Gemini (enforces 5-part format, web search)2102. **codex.md** - System instructions for Codex (enforces repo-aware analysis)2113. **.geminiignore** - Privacy exclusions beyond `.gitignore`2124. **consultation-log-parser.sh** - View consultation history (optional)213214**Installation**: Copy from `~/.claude/skills/multi-ai-consultant/templates/` to project root215216---217218## Privacy & Security219220**Automatic protection**:221- Both CLIs respect `.gitignore` automatically222- Create `.geminiignore` for extra exclusions (`.env*`, `*secret*`, `*credentials*`)223- Pre-consultation check warns if sensitive patterns detected224225**Privacy best practices**:226- Always configure `.gitignore` properly227- Create `.geminiignore` for extra safety228- Use smart context selection (specific files, not entire repo)229- Verify what will be sent: `git status --ignored`230231**For detailed privacy configuration**: Load `references/setup-guide.md` when setting up `.geminiignore` or privacy exclusions.232233---234235## Cost Tracking236237Every consultation logged to `~/.claude/ai-consultations/consultations.log`238239**Log format**: `timestamp,ai,model,input_tokens,output_tokens,cost,project_path`240241**View logs**: `consultation-log-parser.sh --summary`242243**Example output**:244```245Total consultations: 47246Gemini: 23 ($4.25), Codex: 12 ($1.85), Fresh Claude: 12 ($0.00)247Total cost: $6.10248```249250**For detailed cost tracking**: Load `references/commands-reference.md` when viewing logs, calculating costs, or managing budgets.251252---253254## Common Issues255256**For complete troubleshooting**: Load `references/troubleshooting.md` when encountering errors or setup issues.257258**Top 5 issues**:2592601. **CLI not installed**: `gemini: command not found` → Fix: `bun add -g @google/generative-ai-cli`2612. **API key invalid**: Authentication failed → Fix: `export GEMINI_API_KEY="..."`2623. **Context too large**: Token limit exceeded → Fix: Use specific files, not entire repo2634. **Privacy leak**: `.env` file sent → Fix: Add to `.gitignore` and `.geminiignore`2645. **Skill not discovered**: Not working → Fix: Check `~/.claude/skills/multi-ai-consultant`265266---267268## Token Efficiency269270### Without This Skill271272**Typical scenario** (stuck on bug):2731. Try approach 1 (~4k tokens)2742. Research CLI syntax (~3k tokens)2753. Try approach 2 (~4k tokens)2764. Research documentation (~3k tokens)2775. Try approach 3 (~4k tokens)278279**Total**: ~20k tokens, 30-45 minutes280281### With This Skill282283**Same scenario**:2841. Try approach 1 (~4k tokens)2852. Execute `/consult-gemini` (~1k tokens)2863. Gemini finds issue (<5k tokens, billed separately)2874. Implement fix (~3k tokens)288289**Total**: ~8k tokens, 5-10 minutes290291**Savings**: ~60% tokens, ~75% time292293---294295## Success Metrics296297### Time Efficiency298- **Without skill**: 30-45 minutes (trial and error)299- **With skill**: 5-10 minutes (consultation + fix)300- **Savings**: ~75%301302### Token Efficiency303- **Without skill**: ~20k tokens (multiple attempts)304- **With skill**: ~8k tokens (one consultation)305- **Savings**: ~60%306307### Error Prevention308- **Manual CLI use**: 3-5 common errors (flags, parsing, privacy)309- **With skill**: 0 errors (all handled by commands)310- **Prevention**: 100%311312### Quality313- **Manual**: Risk of not synthesizing (just copying external AI)314- **With skill**: Forced synthesis via `GEMINI.md`/`codex.md`315- **Improvement**: Guaranteed value-add316317---318319## Why CLI Approach (Not MCP)?320321| Aspect | MCP Server | CLI Approach |322|--------|------------|--------------|323| Setup time | 4-6 hours | 60-75 minutes |324| Complexity | High (MCP protocol) | Low (bash + CLIs) |325| Maintenance | Update MCP SDK | Update CLI (rare) |326| Flexibility | Locked to AIs | Any AI with CLI |327| Debugging | MCP protocol | Standard bash |328| Dependencies | MCP SDK, npm | Just CLIs |329330**Winner**: CLI approach - 80% less effort, same functionality331332---333334## When to Load References335336Load reference files when working on specific aspects of AI consultation:337338### ai-strengths.md339Load when:340- **Selection-based**: Choosing which AI to consult (Gemini vs Codex vs Fresh Claude)341- **Comparison-based**: Understanding capabilities, costs, and trade-offs between AIs342- **Strategy-based**: Planning combination strategies (free → paid, paid first, budget-conscious)343- **Capability-based**: Understanding special features (Google Search, extended thinking, grounding, repo-aware, fresh perspective)344- **Scenario-based**: Multiple AI consultation workflows, validation workflows345346### setup-guide.md347Load when:348- **Installation-based**: Setting up Gemini CLI, Codex CLI, or installing the skill349- **Configuration-based**: Configuring API keys, environment variables, or system paths350- **Privacy-based**: Setting up .geminiignore, privacy exclusions, or security configurations351- **Template-based**: Installing GEMINI.md, codex.md, .geminiignore, or consultation-log-parser.sh352- **Verification-based**: Testing CLI installation, API keys, or skill discovery353354### commands-reference.md355Load when:356- **Command-based**: Using /consult-gemini, /consult-codex, /consult-claude, or /consult-ai357- **Syntax-based**: Understanding command flags, options, or context selection358- **Cost-based**: Understanding cost tracking, log format, or viewing consultation history359- **Logging-based**: Using consultation-log-parser.sh or analyzing consultation patterns360361### usage-examples.md362Load when:363- **Scenario-based**: Learning how to use skill for bugs, architecture decisions, or code review364- **Workflow-based**: Understanding consultation workflow, synthesis process, or multi-AI approach365- **Example-based**: Seeing real-world examples of consultations and their outcomes (5 detailed examples available)366367### troubleshooting.md368Load when:369- **Error-based**: Encountering specific errors (command not found, API key invalid, parsing failures)370- **Diagnosis-based**: Troubleshooting CLI issues, API connectivity, or skill discovery371- **Fix-based**: Resolving known issues with step-by-step solutions (8 common issues documented)372- **Debugging-based**: Testing CLIs manually, checking configurations, or verifying installations373374---375376## Contributing377378**Found an issue?**379- Document it in troubleshooting.md380- Include fix/workaround381- Update slash commands to prevent382383**Adding new AI?**384- Create new slash command: `commands/consult-newai.md`385- Add to router: Update `commands/consult-ai.md`386- Create template: `templates/newai.md` (if CLI supports system instructions)387- Update documentation388389**Improving synthesis?**390- Edit templates: `templates/GEMINI.md`, `templates/codex.md`391- Test with real consultations392- Measure before/after quality393394---395396## References397398### External Resources399400- **Gemini CLI**: https://ai.google.dev/gemini-api/docs/cli401- **OpenAI Codex**: https://www.npmjs.com/package/codex402- **OpenAI API**: https://platform.openai.com/docs403- **Gemini API Pricing**: https://ai.google.dev/pricing404- **OpenAI Pricing**: https://openai.com/pricing405406### Internal Files407408- **Planning docs**: `planning/multi-ai-consultant-*.md`409- **Slash commands**: `commands/*.md`410- **Templates**: `templates/*`411- **Scripts**: `scripts/*`412- **References**: `references/*.md` (5 reference files)413414---415416## License417418MIT License - See LICENSE file419420---421422**Last Updated**: 2025-11-07423**Status**: Production Ready424**Maintainer**: Claude Skills Maintainers | maintainers@example.com