Skip to content
Back to search

llm-council-skill

A skill to claude code that enables brainstorming with other LLMs (ChatGPT, Gemini) before presenting the implementation plan to the user

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/gcpdev--llm-council-skill/badge.svg)](/listings/gcpdev--llm-council-skill)

HTML

<a href="/listings/gcpdev--llm-council-skill"><img src="/api/listings/gcpdev--llm-council-skill/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

1. **Install the skill** in Claude by uploading the `llm-council.skill` file 2. **Set up API keys and model preferences**: - Copy `.env.template` to create a `.env` file in your working directory - Add your OpenAI API key (get it at https://platform.openai.com/api-keys) - Add your Gemini API key (get it at https://aistudio.google.com/app/apikey) - Optionally customize which models to use (see Model Options below)

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 141 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # LLM Council Skill
  2. A Claude skill that enables collaborative brainstorming with multiple AI models (ChatGPT and Gemini) before presenting implementation plans.
  3. ## What It Does
  4. When you ask Claude to consult with other AI models, Claude will:
  5. 1. Query both ChatGPT and Gemini for their perspectives
  6. <img width="607" height="210" alt="council_demo" src="https://github.com/user-attachments/assets/6e2a759d-40cc-4bc2-9651-a74aca651303" />
  7. 3. Analyze their responses and identify valuable insights
  8. 4. Synthesize a comprehensive implementation plan incorporating ideas from all three models
  9. 5. Present the final plan with attribution to each model's contributions
  10. <img width="608" height="222" alt="council_demo2" src="https://github.com/user-attachments/assets/327893bd-9806-4cfb-85df-538649c978fa" />
  11. ## How to Use
  12. Simply ask Claude to consult with other AI models using phrases like:
  13. - "**Consult the council:** How should I architect a microservices system?"
  14. - "**Ask ChatGPT and Gemini** what they think about my database design"
  15. - "**Get perspectives from other AI models** on this technical decision"
  16. - "**Consult with other LLMs:** What's the best approach for..."
  17. <img width="609" height="79" alt="council_demo3" src="https://github.com/user-attachments/assets/ba98d179-08be-4cc8-ac6a-0e1bc69b3fc7" />
  18. **Example:**
  19. ```
  20. User: Consult the council: How should I structure my React app for scalability?
  21. Claude will then:
  22. - Query ChatGPT and Gemini about React architecture
  23. - Analyze their suggestions on components, state management, and organization
  24. - Present a synthesized plan incorporating insights from all three models
  25. ```
  26. ## Installation
  27. 1. **Install the skill** in Claude by uploading the `llm-council.skill` file
  28. 2. **Set up API keys and model preferences**:
  29. - Copy `.env.template` to create a `.env` file in your working directory
  30. - Add your OpenAI API key (get it at https://platform.openai.com/api-keys)
  31. - Add your Gemini API key (get it at https://aistudio.google.com/app/apikey)
  32. - Optionally customize which models to use (see Model Options below)
  33. ## Model Options
  34. **Default Models (Fast & Cost-Effective):**
  35. - ChatGPT: `gpt-5-nano-2025-08-07` (highly cost-effective)
  36. - Gemini: `gemini-3-flash-preview` (balanced speed and intelligence)
  37. **Upgrade Options for Better Collaboration:**
  38. **OpenAI models (ordered by capability):**
  39. - `gpt-5-nano` - Fastest, cheapest version of GPT-5. It's great for summarization and classification tasks. (Default)
  40. - `gpt-5-mini` - Balanced cost and quality
  41. - `gpt-5.2` - Smart model, capable of most tasks
  42. - `gpt-5.2-pro` - State-of-the-art for professional knowledge work
  43. **Gemini models (ordered by capability):**
  44. - `gemini-2.5-flash-lite` - Ultra-fast, optimized for throughput
  45. - `gemini-2.5-flash` - Best price-performance
  46. - `gemini-3-flash-preview` - Balanced (default)
  47. - `gemini-3-pro-preview` - Most intelligent, best reasoning
  48. **How to Configure:**
  49. Add these lines to your `.env` file:
  50. ```
  51. OPENAI_MODEL=gpt-5-nano
  52. GEMINI_MODEL=gemini-3-flash-preview
  53. ```
  54. **Recommended Configurations:**
  55. - **Balanced**: Defaults (`gpt-5-nano` + `gemini-3-flash-preview`)
  56. - **Budget**: `gpt-5-nano` + `gemini-2.5-flash`
  57. - **High Quality**: `gpt-5` + `gemini-3-flash-preview`
  58. - **Premium Reasoning**: `gpt-5.2` + `gemini-3-pro-preview`
  59. - **Professional Work**: `gpt-5.2-pro` + `gemini-3-pro-preview`
  60. ## Benefits
  61. - **Diverse perspectives**: Get insights from three different AI models with different training and capabilities
  62. - **Better decisions**: Identify potential issues or alternatives you might have missed with a single model
  63. - **Comprehensive planning**: Combine strengths of multiple models for more robust implementation plans
  64. ## API Costs
  65. Both OpenAI and Gemini APIs have usage costs that vary significantly by model:
  66. **OpenAI Cost Tiers (approximate, check current pricing):**
  67. *Budget Tier:*
  68. - `gpt-5-nano`: Very low cost per token
  69. -
  70. *Standard Tier:*
  71. - `gpt-5-mini`: Moderate cost
  72. *Premium Tier:*
  73. - `gpt-5.2`: Higher cost per token
  74. - `gpt-5.2-pro`: Highest cost for professional work
  75. **Gemini Cost Tiers (approximate, check current pricing):**
  76. *Budget Tier:*
  77. - `gemini-2.5-flash-lite`: Very low cost, optimized for throughput
  78. - `gemini-2.5-flash`: Low cost, best price-performance
  79. *Standard Tier:*
  80. - `gemini-3-flash-preview`: Moderate cost (default)
  81. *Premium Tier:*
  82. - `gemini-3-pro-preview`: Higher cost for advanced reasoning
  83. **Cost Management Tips:**
  84. - Start with default models for routine brainstorming (very cost-effective)
  85. - Use mid-tier models (`gpt-5-mini` + `gemini-3-flash-preview`) for balanced quality/cost
  86. - Upgrade to premium models only for critical architectural decisions or complex reasoning tasks
  87. - Set usage limits in your API dashboards (OpenAI Platform and Google AI Studio)
  88. - Consider setting monthly budgets to avoid surprises
  89. - Monitor your usage patterns and adjust model choices accordingly
  90. **Note:** Each `/council` command makes 2 API calls (one to ChatGPT, one to Gemini), so total cost is the sum of both models' pricing.
  91. ## Skill Structure
  92. ```
  93. llm-council/
  94. ├── SKILL.md # Main skill instructions
  95. ├── scripts/
  96. │ └── query_llms.py # Python script that queries both APIs
  97. └── references/
  98. └── setup.md # Detailed setup instructions
  99. ```
  100. ## Troubleshooting
  101. **"API key not found" error**: Make sure your `.env` file is in the current working directory with the correct keys.
  102. **API timeout**: The script has a 30-second timeout per API. If an API is slow or down, it will show an error but continue with the other model's response.
  103. **One API fails**: Claude will note which model's perspective is unavailable and proceed with the available responses.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.