An MCP server that interfaces with OpenAI, Google, and Anthropic's APIs to give Claude Code "coworkers" to help it on difficult problems.
A Model Context Protocol (MCP) server that allows Claude Code to consult with additional AI agents for code and problem analysis. This server provides access to Darren (OpenAI), Sonny (Anthropic), Sergey (OpenAI with web search), and Gemma (Google Gemini with repository analysis) as expert consultants, enabling multi-model perspective on coding problems.
Clone the repository:
git clone https://github.com/yourusername/consulting-agents-mcp.git
cd consulting-agents-mcp
Create and activate a virtual environment:
python -m venv mcp_venv
source mcp_venv/bin/activate # On Windows: mcp_venv\Scripts\activate
Install dependencies:
pip install -r requirements.txt
Set up API keys:
Create a .env file in the project root:
OPENAI_API_KEY=your_openai_api_key_here
ANTHROPIC_API_KEY=your_anthropic_api_key_here
GOOGLE_API_KEY=your_google_api_key_here
Start the server:
chmod +x start_mcp_server.sh
./start_mcp_server.sh
Register the MCP server with Claude Code:
claude mcp add ConsultingAgents /absolute/path/to/consulting-agents-mcp/start_mcp_server.sh
Start Claude Code with MCP integration:
claude --mcp-debug
Use the tools in Claude Code:
Now you can use consult_with_darren, consult_with_sonny, consult_with_sergey, and consult_with_gemma functions in Claude Code.
The MCP server provides four consulting tools:
consult_with_darrenUses OpenAI’s o3-mini model with high reasoning to analyze code and provide recommendations.
Parameters:
consultation_context: Description of the problem (required)source_code: Optional code to analyzeconsult_with_sonnyUses Claude 3.7 Sonnet with enhanced thinking to provide in-depth code analysis.
Parameters:
consultation_context: Description of the problem (required)source_code: Optional code to analyzeNote: This agent is somewhat redundant now that Claude Code has native Extended Thinking mode, but may still be useful for getting a second opinion or different approach from another Claude model.
consult_with_sergeyUses GPT-4o with web search capabilities to find relevant documentation and examples.
Parameters:
consultation_context: Description of what information or documentation you need (required)search_query: Optional specific search query to usesource_code: Optional code for contextNote: This agent is somewhat redundant now that Claude Code has native web search capabilities, but may still be useful for comparing search results between GPT-4o and Claude, or getting a different perspective.
consult_with_gemmaUses Google’s Gemini 2.5 Pro model with 1M token context window to analyze entire repositories and provide comprehensive development plans.
Parameters:
consultation_context: Description of the task or feature to be implemented (required)repo_url: GitHub repository URL to analyze (required) - IMPORTANT: Always specify the complete and correct GitHub URL (e.g., “https://github.com/username/repo”)feature_description: Detailed description of the feature to implement (required)Note: This agent is particularly useful as Claude Code does not natively have the ability to analyze entire repositories in a single context.
Important URL Specification: When using Gemma, always provide the exact GitHub repository URL. Claude Code may incorrectly infer the repository URL from your local directory path, which can lead to repository access errors. The URL should be in the format https://github.com/username/repository with the correct case sensitivity.
MCP_TRANSPORT: Transport protocol (default: “stdio”, alternatives: “http”, “sse”)HOST: Server host when using HTTP/SSE transport (default: “127.0.0.1”)PORT: Server port when using HTTP/SSE transport (default: 5000)When running with HTTP transport, the server provides these endpoints:
GET /health
Returns server status and available agents.
POST /consult
Request body for Darren or Sonny:
{
"agent": "Darren",
"consultation_context": "I have a bug in my code where...",
"source_code": "def example():\n return 'hello'"
}
Request body for Sergey:
{
"agent": "Sergey",
"consultation_context": "How do I implement JWT authentication in Express?",
"search_query": "express.js JWT auth implementation"
}
Request body for Gemma:
{
"agent": "Gemma",
"consultation_context": "Adding user authentication to the API",
"repo_url": "https://github.com/username/repo",
"feature_description": "Implement basic username/password authentication for API access"
}
Important: Always provide the exact and complete GitHub repository URL in the repo_url field. Do not rely on Claude Code to infer this from your local directory path.
When updating to a new version of consulting-agents-mcp, follow these steps:
./start_mcp_server.sh
claude mcp remove ConsultingAgents
claude mcp add ConsultingAgents /absolute/path/to/consulting-agents-mcp/start_mcp_server.sh
This process ensures Claude Code is using the updated version of the MCP server with any new models or functionality.
Start the server with debug output:
DEBUG=true ./start_mcp_server.sh
Test HTTP endpoints (when using HTTP transport):
# Test Darren
curl -X POST http://localhost:5000/consult \
-H "Content-Type: application/json" \
-d '{"agent":"Darren","consultation_context":"Test message"}'
# Test Sonny
curl -X POST http://localhost:5000/consult \
-H "Content-Type: application/json" \
-d '{"agent":"Sonny","consultation_context":"Test message"}'
# Test Sergey
curl -X POST http://localhost:5000/consult \
-H "Content-Type: application/json" \
-d '{"agent":"Sergey","consultation_context":"Test message","search_query":"example"}'
# Test Gemma
curl -X POST http://localhost:5000/consult \
-H "Content-Type: application/json" \
-d '{"agent":"Gemma","consultation_context":"Add user authentication","repo_url":"https://github.com/username/repo","feature_description":"Implement basic username/password authentication for API access"}'
mcp_consul_server.py: Main MCP server implementationstart_mcp_server.sh: Script to start the server with proper environmentrequirements.txt: Python dependenciesMIT
Contributions are welcome! Please feel free to submit a Pull Request.
Mcp2tavily
一个MCP服务器，让你的Claude Cline以及Langchain实现网络搜索功能。An MCP server that allows your Claude Cline and Langchain to implement network search functions.
Juliadoc Mcp
MCP server for efficiently retrieving Julia docstrings and source code
Mcp Notion
Global Notion workspace-accessible MCP server for all Notion pages within the workspace
A Python server implementation for WeCom (WeChat Work) bot that follows the Model Context Protocol (MCP). This server provides a standardized interface for handling automated messaging and context-aware interactions within enterprise WeChat environments.
