Phone-a-Friend MCP Server 🧠📞
About
# Phone-a-Friend MCP Server 🧠📞 An AI-to-AI consultation system that enables one AI to "phone a friend" (another AI) for critical thinking, long context reasoning, and complex problem solving via OpenRouter. ## The Problem 🤔 Sometimes an AI encounters complex problems that require: - **Deep critical thinking** beyond…
Details
- License
- MIT license
Explore
- Two-step consultation: context + reasoning then extract actionable insights.
- Automatic model selection based on provider and task.
- Supports OpenRouter, OpenAI, Anthropic, and Google/Gemini.
- Exposes a single phone_a_friend MCP tool.
- Handles long context reasoning (>100k tokens).
- No memory of past conversations; quality depends on provided context.
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Phone-a-Friend MCP Server 🧠📞Command (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Option 1: Using uv (Recommended)
{
"mcpServers": {
"phone-a-friend": {
"command": "uvx",
"args": [
"run",
"--refresh",
"phone-a-friend-mcp-server",
],
"env": {
"OPENROUTER_API_KEY": "your-openrouter-api-key",
"PHONE_A_FRIEND_MODEL": "anthropic/claude-4-opus"
}
}
}
}
1. Clone the repository:
git clone https://github.com/abhishekbhakat/phone-a-friend-mcp-server.git
cd phone-a-friend-mcp-server
2. Install dependencies:
uv pip install -e .
3. Configure your preferred AI provider:
OpenRouter (recommended - access to multiple models):
export OPENROUTER_API_KEY="your-openrouter-key"
To use Phone-a-Friend MCP server with Claude Desktop, add this configuration to your claude_desktop_config.json file:
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%\Claude\claude_desktop_config.json
You can configure different AI providers directly in the Claude Desktop config:
{
"mcpServers": {
"phone-a-friend": {
"command": "phone-a-friend-mcp-server",
"env": {
"OPENROUTER_API_KEY": "your-openrouter-api-key",
"PHONE_A_FRIEND_MODEL": "anthropic/claude-4-opus"
}
}
}
}
Alternative Providers:
{
"mcpServers": {
"phone-a-friend-openai": {
"command": "phone-a-friend-mcp-server",
"env": {
"OPENAI_API_KEY": "your-openai-api-key"
}
},
"phone-a-friend-anthropic": {
"command": "phone-a-friend-mcp-server",
"env": {
"ANTHROPIC_API_KEY": "your-anthropic-api-key"
}
},
"phone-a-friend-google": {
"command": "phone-a-friend-mcp-server",
"env": {
"GOOGLE_API_KEY": "your-google-api-key"
}
}
}
}
1. Install Phone-a-Friend MCP Server (see Installation section above)
2. Open Claude Desktop Settings → Developer → Edit Config
3. Add the configuration (choose one of the options above)
4. Replace paths and API keys with your actual values
5. Restart Claude Desktop
6. Look for the 🔨 hammer icon in the input box to confirm the server is connected
pytest
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"phone-a-friend mcp server \ud83e\udde0\ud83d\udcde": {
"phone-a-friend-mcp-server": {
"command": "uv",
"args": [
"pip",
"install",
"-e",
"."
]
}
}
}
}McpServers
{
"phone-a-friend-mcp-server": {
"command": "uv",
"args": [
"pip",
"install",
"-e",
"."
]
}
}An AI-to-AI consultation system that enables one AI to "phone a friend" (another AI) for critical thinking, long context reasoning, and complex problem solving via OpenRouter.
The Problem 🤔
Sometimes an AI encounters complex problems that require:
- Deep critical thinking beyond immediate capabilities
- Long context reasoning with extensive information
- Multi-step analysis that benefits from external perspective
- Specialized expertise from different AI models
The Solution �
Phone-a-Friend MCP Server creates a two-step consultation process:
1. Context + Reasoning: Package all relevant context and send to external AI for deep analysis
2. Extract Actionable Insights: Process the reasoning response into usable format for the primary AI
This enables AI systems to leverage other AI models as "consultants" for complex reasoning tasks.
Architecture 🏗️
Primary AI → Phone-a-Friend MCP → OpenRouter → External AI (GPT-4, Claude, etc.) → Processed Response → Primary AI
Sequential Workflow:
1. analyze_context - Gather and structure all relevant context
2. get_critical_thinking - Send context to external AI via OpenRouter for reasoning
3. extract_actionable_insights - Process response into actionable format
When to Use 🎯
Ideal for:
- Complex multi-step problems requiring deep analysis
- Situations needing long context reasoning (>100k tokens)
- Cross-domain expertise consultation
- Critical decision-making with high stakes
- Problems requiring multiple perspectives
Not needed for:
- Simple factual questions
- Basic reasoning tasks
- Quick responses
- Well-defined procedural tasks
Installation 🚀
1. Clone the repository:
git clone https://github.com/abhishekbhakat/phone-a-friend-mcp-server.git
cd phone-a-friend-mcp-server
2. Install dependencies:
uv pip install -e .
3. Configure your preferred AI provider:
OpenRouter (recommended - access to multiple models):
```bash
export OPENROUTER_API_KEY="your-openrouter-key"
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
