Perplexity Server
MCP web search using perplexity without any API KEYS
wysh3
README
Perplexity MCP Zerver <a href="https://raw.githubusercontent.com/wysh3/perplexity-mcp-zerver/main/README.md" title="Copy Full README Content (opens raw file view)">📋</a>
A research level Model Context Protocol (MCP) server implementation providing AI-powered research capabilities by interacting with the Perplexity website without requiring an API key.
Features
- 🔍 Web search integration via Perplexity's web interface.
- 💬 Persistent chat history for conversational context.
- 📄 Tools for documentation retrieval, API finding, and code analysis.
- 🚫 No API Key required (relies on web interaction).
- 🛠️ TypeScript-first implementation.
- 🌐 Uses Puppeteer for browser automation.
Tools
1. Search (search
)
Performs a search query on Perplexity.ai. Supports brief
, normal
, or detailed
responses. Returns raw text output.
2. Get Documentation (get_documentation
)
Asks Perplexity to provide documentation and examples for a technology/library, optionally focusing on specific context. Returns raw text output.
3. Find APIs (find_apis
)
Asks Perplexity to find and evaluate APIs based on requirements and context. Returns raw text output.
4. Check Deprecated Code (check_deprecated_code
)
Asks Perplexity to analyze a code snippet for deprecated features within a specific technology context. Returns raw text output.
5. Extract URL Content (extract_url_content
)
Extracts main article text content from URLs using browser automation and Mozilla's Readability. Handles GitHub repositories via gitingest.com. Supports recursive link exploration up to depth. Returns structured JSON with content and metadata.
6. Chat (chat_perplexity
)
Maintains ongoing conversations with Perplexity AI. Stores chat history locally in chat_history.db
within the project directory. Returns a stringified JSON object containing chat_id
and response
.
Installation
just copy <a href="https://raw.githubusercontent.com/wysh3/perplexity-mcp-zerver/main/README.md" title="Copy Full README Content (opens raw file view)">📋</a> and paste the readme and let the AI take care of the rest
- Clone or download this repository:
git clone https://github.com/wysh3/perplexity-mcp-zerver.git
cd perplexity-mcp-zerver
- Install dependencies:
npm install
- Build the server:
npm run build
Important: Ensure you have Node.js installed. Puppeteer will download a compatible browser version if needed during installation. Restart your IDE/Application after building and configuring the project for changes to take effect.
Configuration
Add the server to your MCP configuration file (e.g., cline_mcp_settings.json
for the VS Code extension or claude_desktop_config.json
for the desktop app).
Important: Replace /path/to/perplexity-mcp-zerver/build/index.js
with the absolute path to the built index.js
file on your system.
Example for Cline/RooCode Extension:
{
"mcpServers": {
"perplexity-server": {
"command": "node",
"args": [
"/full/path/to/your/perplexity-mcp-zerver/build/index.js" // <-- Replace this path! (in case of windows for ex: "C:\\Users\\$USER\\Documents\\Cline\\MCP\\perplexity-mcp-zerver\\build\\index.js"
],
"env": {},
"disabled": false,
"alwaysAllow": [],
"autoApprove": [],
"timeout": 300
}
}
}
Example for Claude Desktop:
{
"mcpServers": {
"perplexity-server": {
"command": "node",
"args": [
"/full/path/to/your/perplexity-mcp-zerver/build/index.js" // <-- Replace this path!
],
"env": {},
"disabled": false,
"alwaysAllow": []
}
}
}
Usage
- Ensure the server is configured correctly in your MCP settings file.
- Restart your IDE (like VS Code with the Cline/RooCode extension) or the Claude Desktop application.
- The MCP client should automatically connect to the server.
- You can now ask the connected AI assistant (like Claude) to use the tools, e.g.:
- "Use perplexity-server search to find the latest news on AI."
- "Ask perplexity-server get_documentation about React hooks."
- "Start a chat with perplexity-server about quantum computing."
Credits
Thanks DaInfernalCoder:
License
This project is licensed under the GNU General Public License v3.0 - see the LICENSE.md file for details. Portions are derived from MIT-licensed work, as noted in the license file.
Disclaimer
This project interacts with the Perplexity website via web automation (Puppeteer). It is intended for educational and research purposes only. Web scraping and automation may be against the terms of service of the target website. The author does not endorse or encourage any unauthorized automation or violation of terms of service. Use responsibly and ethically. The stability of this server depends on the Perplexity website's structure remaining consistent.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
AIO-MCP Server
🚀 All-in-one MCP server with AI search, RAG, and multi-service integrations (GitLab/Jira/Confluence/YouTube) for AI-enhanced development workflows. Folk from
Persistent Knowledge Graph
An implementation of persistent memory for Claude using a local knowledge graph, allowing the AI to remember information about users across conversations with customizable storage location.
Hyperbrowser MCP Server
Welcome to Hyperbrowser, the Internet for AI. Hyperbrowser is the next-generation platform empowering AI agents and enabling effortless, scalable browser automation. Built specifically for AI developers, it eliminates the headaches of local infrastructure and performance bottlenecks, allowing you to
React MCP
react-mcp integrates with Claude Desktop, enabling the creation and modification of React apps based on user prompts
Atlassian Integration
Model Context Protocol (MCP) server for Atlassian Cloud products (Confluence and Jira). This integration is designed specifically for Atlassian Cloud instances and does not support Atlassian Server or Data Center deployments.