Perplexity Insight MCP Server
An MCP server implementation for interfacing with the Perplexity AI API, providing advanced question answering capabilities through the standardised Model Context Protocol.
Features
- Seamless integration with Perplexity AI API
- Support for different Perplexity models (sonar-reasoning, sonar-pro, sonar-deep-research)
- Customisable system prompts and user queries
- Proper error handling and response formatting
- Rate limiting protection
- Easy integration with Windsurf IDE
Requirements
- Node.js 18+
- Perplexity API key
Installation
Environment Variables
Create a .env
file with the following variables:
Usage
Run the server:
API Tools
The server exposes the following tools:
perplexity_ask
- Send a direct question to Perplexity AIperplexity_search
- Perform a search query with Perplexity AI
Changing Models
Both tools support the following Perplexity models:
sonar-reasoning
(default) - Perplexity's reasoning-focused model, best for general questionssonar-pro
- Enhanced model with improved capabilities for professional use casessonar-deep-research
- Specialised for in-depth research and complex queries
To specify a model when using the tools, include the model
parameter in your request:
You can also customise the system prompt and maximum token count:
Tool Response Format
The server follows the MCP specification for tool responses:
Windsurf Integration
Setting up in Windsurf
- Build the server:
- Open Windsurf and navigate to Settings
- Find the "AI Settings" or "Model Context Protocol" section
- Add a new MCP server with the following details:
- Name: Perplexity Insight
- Type: Local Process
- Command: Path to your Node.js executable
- Arguments: Path to your compiled
index.js
file - Working Directory: Path to your project directory
- Environment Variables: Make sure to include
PERPLEXITY_API_KEY=your_api_key_here
- Enable the server and restart Windsurf if necessary
Example Configuration
Here's an example configuration for the mcp_config.json
file:
Replace /path/to/perplexity-insight-MCP
with the actual path to your installation directory and use your actual Perplexity API key.
Using Perplexity in Windsurf
- Use the AI Assistant panel to ask questions that will be directed to Perplexity
- For web searches, include specific terms like "search for" in your queries
- To change models, include the model name in your query as shown in the "Changing Models" section
- Windsurf will automatically use the appropriate Perplexity tool based on your query
Development
For local development:
Troubleshooting
If you encounter issues with the MCP server:
- Check that your API key is valid and properly set in the
.env
file - Verify that the response format matches the MCP specification
- Look for any error messages in the server logs
- Ensure Windsurf is properly configured to use the MCP server
License
MIT
This server cannot be installed
remote-capable server
The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.
Interfaces with the Perplexity AI API to provide advanced question answering capabilities through the standardized Model Context Protocol, supporting multiple Perplexity models.
Related MCP Servers
- AsecurityFlicenseAqualityThis server provides access to the Perplexity AI API, enabling interaction through chatting, searching, and documentation retrieval within MCP-based systems.Last updated -52JavaScript
- AsecurityFlicenseAqualityProvides a standardized way to integrate Perplexity AI's features like chat, search, and documentation access into MCP-based systems.Last updated -5JavaScript
- AsecurityFlicenseAqualityA Node.js implementation that enables Claude to interact with Perplexity AI's language models through Anthropic's Model Context Protocol, providing tools for advanced chat completions and quick queries.Last updated -26TypeScript
- -securityAlicense-qualityProvides access to Perplexity AI models through two tools: ask\_perplexity for expert programming assistance and chat\_perplexity for maintaining ongoing conversations with context preservation.Last updated -PythonMIT License