perplexity-mcp
A Model Context Protocol (MCP) server for research and documentation assistance using Perplexity AI
Documentation
MCP-researcher Server
Your own research assistant inside of Claude! Utilizes Perplexity's Sonar Pro API to get documentation, create up-to-date API routes, and check deprecated code. Includes Chain of Thought Reasoning and local chat history through SQLite.
Features
1. Search
Performs general search queries to get comprehensive information on any topic. Supports different detail levels (brief, normal, detailed) to get tailored responses.
2. Get Documentation
Retrieves documentation and usage examples for specific technologies, libraries, or APIs. Get comprehensive documentation including best practices and common pitfalls.
3. Find APIs
Discovers and evaluates APIs that could be integrated into a project. Get detailed analysis of features, pricing, and integration complexity.
4. Check Deprecated Code
Analyzes code for deprecated features or patterns, providing migration guidance. Helps modernize code by suggesting current best practices.
Prerequisites
1. System Requirements:
# On macOS with Homebrew:
brew install python-setuptools2. API Key:
Installation
1. Create Project Directory:
mkdir -p ~/Documents/Claude/MCP
cd ~/Documents/Claude/MCP2. Clone the Repository:
git clone https://github.com/pashpashpash/perplexity-mcp.git
cd perplexity-mcp3. Install Dependencies:
npm installNote: If you see Python distutils errors, make sure you've installed python-setuptools as mentioned in prerequisites.
4. Build the Project:
npm run buildThis will create the build directory with the compiled server code.
Configuration
1. Configure Claude Desktop:
Add this to your claude_desktop_config.json:
- macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`
- Windows: `%APPDATA%/Claude/claude_desktop_config.json`
{
"mcpServers": {
"perplexity-server": {
"command": "node",
"args": ["path/to/perplexity-mcp/build/index.js"],
"env": {
"PERPLEXITY_API_KEY": "your-api-key-here"
},
"autoApprove": [
"search",
"get_documentation",
"find_apis",
"check_deprecated_code"
]
}
}
}Note:
- Replace "path/to/perplexity-mcp" with the absolute path to your cloned repository
- Replace "your-api-key-here" with your Perplexity API key
- Make sure to use "/build/index.js" (not "/dist/index.js")
Starting the Server
1. Manual Start:
cd path/to/perplexity-mcp
PERPLEXITY_API_KEY="your-api-key-here" node build/index.js2. Verify Server:
The server should start without any errors. Keep this terminal window open while using the server.
Example Usage
Search
{
"query": "What are the best practices for React hooks?",
"detail_level": "detailed"
}Get Documentation
{
"technology": "React",
"topic": "useEffect hook",
"include_examples": true
}Find APIs
{
"category": "payment processing",
"requirements": ["recurring billing", "international support"]
}Check Deprecated Code
{
"code": "class MyComponent extends React.Component {...}",
"framework": "React",
"version": "18"
}Troubleshooting
1. Build Directory Issues:
2. Server Connection Issues:
tail -n 20 -f ~/Library/Logs/Claude/mcp*.log3. Python Dependencies:
brew install python-setuptoolsDevelopment
# Install dependencies
npm install
# Build the project
npm run build
# Development with auto-rebuild
npm run watch
# Start server with debug output
DEBUG=* node build/index.jsDocumentation
For detailed examples and usage guides, see:
License
MIT
Note: This is a fork of the original perplexity-mcp repository.
Frequently asked questions
What is perplexity-mcp?
perplexity-mcp is A Model Context Protocol (MCP) server for research and documentation assistance using Perplexity AI
How do I install perplexity-mcp?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is perplexity-mcp open source?
Yes — it is hosted on GitHub at https://github.com/pashpashpash/perplexity-mcp and has 38 stars.
Related MCP tools
🔥 Official Firecrawl MCP Server - Adds powerful web scraping and search to Cursor, Claude and any other LLM clients.
Use any LLMs (Large Language Models) for Deep Research. Support SSE API and MCP server.
Enhanced MCP server for interactive user feedback and command execution in AI-assisted development, featuring dual interface support (Web UI and Desktop Application) with intelligent environment detection and cross-platform compatibility.
A powerful Zotero AI and MCP plugin with ChatGPT, Gemini 3.7, Claude Fable 5, Claude Opus 5, DeepSeek V4, Grok, OpenRouter, Kimi k3, GLM 5.3, SiliconFlow, GPT-oss, Gemma 4, Qwen 3.8
Connect your browser to AI models. Just use Dia on Chrome, Arc or Firefox.
文颜 MCP Server 可以让 AI 自动将 Markdown 文章排版后发布至微信公众号。
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP