Skip to content
Back to search

cognition-wheel

A Model Context Protocol (MCP) server that implements a "wisdom of crowds" approach to AI reasoning by consulting multiple state-of-the-art language models in parallel and synthesizing their responses.

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/hormold--cognition-wheel/badge.svg)](/listings/hormold--cognition-wheel)

HTML

<a href="/listings/hormold--cognition-wheel"><img src="/api/listings/hormold--cognition-wheel/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.

Option 1: Use with npx (Recommended)

# Run directly with npx (no installation needed)
npx mcp-cognition-wheel

# Or install globally
npm install -g mcp-cognition-wheel
mcp-cognition-wheel

Option 2: Build from source

1. Clone the repository 2. Install dependencies: `pnpm install` 3. Copy `.env.example` to `.env` and add your API keys 4. Build the project: `pnpm run build`

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 · 172 lines

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

  1. # Cognition Wheel MCP Server
  2. A Model Context Protocol (MCP) server that implements a "wisdom of crowds" approach to AI reasoning by consulting multiple state-of-the-art language models in parallel and synthesizing their responses.
  3. ## Quick Start
  4. ### Option 1: Use with npx (Recommended)
  5. ```bash
  6. # Run directly with npx (no installation needed)
  7. npx mcp-cognition-wheel
  8. # Or install globally
  9. npm install -g mcp-cognition-wheel
  10. mcp-cognition-wheel
  11. ```
  12. ### Option 2: Build from source
  13. 1. Clone the repository
  14. 2. Install dependencies: `pnpm install`
  15. 3. Copy `.env.example` to `.env` and add your API keys
  16. 4. Build the project: `pnpm run build`
  17. ## How It Works
  18. The Cognition Wheel follows a three-phase process:
  19. 1. **Parallel Consultation**: Simultaneously queries three different AI models:
  20. - Claude-4-Opus (Anthropic)
  21. - Gemini-2.5-Pro (Google)
  22. - O3 (OpenAI)
  23. 2. **Anonymous Analysis**: Uses code names (Alpha, Beta, Gamma) to eliminate bias during the synthesis phase
  24. 3. **Smart Synthesis**: Randomly selects one of the models to act as a synthesizer, which analyzes all responses and produces a final, comprehensive answer
  25. ## Features
  26. - **Parallel Processing**: All models are queried simultaneously for faster results
  27. - **Bias Reduction**: Anonymous code names prevent synthesizer bias toward specific models
  28. - **Internet Search**: Optional web search capabilities for all models
  29. - **Detailed Logging**: Comprehensive debug logs for transparency and troubleshooting
  30. - **Robust Error Handling**: Graceful degradation when individual models fail
  31. ## Installation
  32. ### Option 1: Use with npx (Recommended)
  33. ```bash
  34. # Run directly with npx (no installation needed)
  35. npx mcp-cognition-wheel
  36. # Or install globally
  37. npm install -g mcp-cognition-wheel
  38. mcp-cognition-wheel
  39. ```
  40. ### Option 2: Build from source
  41. 1. Clone the repository
  42. 2. Install dependencies: `pnpm install`
  43. 3. Copy `.env.example` to `.env` and add your API keys
  44. 4. Build the project: `pnpm run build`
  45. ## Usage
  46. This is an MCP server designed to be used with MCP-compatible clients like Claude Desktop or other MCP tools.
  47. ### Required Environment Variables
  48. - `ANTHROPIC_API_KEY`: Your Anthropic API key
  49. - `GOOGLE_GENERATIVE_AI_API_KEY`: Your Google AI API key
  50. - `OPENAI_API_KEY`: Your OpenAI API key
  51. ### Using with Cursor
  52. Based on the guide from [this dev.to article](https://dev.to/andyrewlee/use-your-own-mcp-on-cursor-in-5-minutes-1ag4), here's how to integrate with Cursor:
  53. #### Option 1: Using npx (Recommended)
  54. 1. **Open Cursor Settings**:
  55. - Go to Settings → MCP
  56. - Click "Add new MCP server"
  57. 2. **Configure the server**:
  58. - **Name**: `cognition-wheel`
  59. - **Command**: `npx`
  60. - **Args**: `["-y", "mcp-cognition-wheel"]`
  61. Example configuration:
  62. ```json
  63. {
  64. "cognition-wheel": {
  65. "command": "npx",
  66. "args": ["-y", "mcp-cognition-wheel"],
  67. "env": {
  68. "ANTHROPIC_API_KEY": "your_anthropic_key",
  69. "GOOGLE_GENERATIVE_AI_API_KEY": "your_google_key",
  70. "OPENAI_API_KEY": "your_openai_key"
  71. }
  72. }
  73. }
  74. ```
  75. #### Option 2: Using local build
  76. 1. **Build the project** (if not already done):
  77. ```bash
  78. pnpm run build
  79. ```
  80. 2. **Configure the server**:
  81. - **Name**: `cognition-wheel`
  82. - **Command**: `node`
  83. - **Args**: `["/absolute/path/to/your/cognition-wheel/dist/app.js"]`
  84. Example configuration:
  85. ```json
  86. {
  87. "cognition-wheel": {
  88. "command": "node",
  89. "args": [
  90. "/Users/yourname/path/to/cognition-wheel/dist/app.js"
  91. ],
  92. "env": {
  93. "ANTHROPIC_API_KEY": "your_anthropic_key",
  94. "GOOGLE_GENERATIVE_AI_API_KEY": "your_google_key",
  95. "OPENAI_API_KEY": "your_openai_key"
  96. }
  97. }
  98. }
  99. ```
  100. 3. **Test the integration**:
  101. - Enter Agent mode in Cursor
  102. - Ask a complex question that would benefit from multiple AI perspectives
  103. - The `cognition_wheel` tool should be automatically triggered
  104. ### Tool Usage
  105. The server provides a single tool called `cognition_wheel` with the following parameters:
  106. - `context`: Background information and context for the problem
  107. - `question`: The specific question you want answered
  108. - `enable_internet_search`: Boolean flag to enable web search capabilities
  109. ## Development
  110. - `pnpm run dev`: Watch mode for development
  111. - `pnpm run build`: Build the TypeScript code
  112. - `pnpm run start`: Run the server directly with tsx
  113. ## Docker
  114. Build and run with Docker:
  115. ```bash
  116. # Build the image
  117. docker build -t cognition-wheel .
  118. # Run with environment variables
  119. docker run --rm \
  120. -e ANTHROPIC_API_KEY=your_key \
  121. -e GOOGLE_GENERATIVE_AI_API_KEY=your_key \
  122. -e OPENAI_API_KEY=your_key \
  123. cognition-wheel
  124. ```
  125. ## License
  126. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

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