Perplexity MCP Server
@DaInfernalCoder
Perplexity MCP Server について
A Model Context Protocol (MCP) server for research and documentation assistance using Perplexity AI. Won 1st @ Cline Hackathon
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"perplexity": {
"command": "node",
"args": [
"/path/to/perplexity-server/build/index.js"
],
"env": {
"PERPLEXITY_API_KEY": "YOUR_API_KEY_HERE"
},
"disabled": false,
"autoApprove": []
}
}
}ツール
ツールは検出されませんでした
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
概要
What is Perplexity MCP Server?
An intelligent research assistant that integrates with Perplexity’s specialized AI models via the Model Context Protocol. It automatically detects query complexity and routes requests to the most appropriate model for optimal results, and forces the AI agent to be specific. Unlike the official server, it provides search capabilities for every task.
How to use Perplexity MCP Server?
Install Node.js, get a Perplexity API key from perplexity.ai/settings/api, then configure your MCP client with the server either by pointing to a local build (node /path/to/build/index.js) or directly via npx -y perplexity-mcp. Pass the API key through the environment variable PERPLEXITY_API_KEY, a command‑line --api-key argument, or a .env file using the --cwd argument. Use the three provided tools: search, reason, and deep_research.
Key features of Perplexity MCP Server
- Three tools: search, reason, deep research.
- Automatic query complexity detection routes to the best model.
- Forces the AI agent to be specific.
- Supports Sonar Pro, Sonar Reasoning Pro, and Sonar Deep Research models.
- Configurable via environment variable, CLI argument, or .env file.
- Works with any MCP‑compatible client.
Use cases of Perplexity MCP Server
- Quick fact lookup using the
searchtool. - Multi‑step analysis and comparisons with the
reasontool. - In‑depth research with comprehensive report generation via
deep_research. - Embedding precise, research‑backed answers into AI assistant workflows.
FAQ from Perplexity MCP Server
What tools does Perplexity MCP Server provide?
search (Sonar Pro) for simple queries, reason (Sonar Reasoning Pro) for complex analysis, and deep_research (Sonar Deep Research) for thorough investigations. Each accepts an optional force_model parameter.
How is this different from the official Perplexity MCP server?
This server provides search capabilities for every task, whereas the official server does not. It also forces the agent to be specific.
What are the runtime and dependency requirements?
Node.js from nodejs.org and a valid Perplexity API key from perplexity.ai/settings/api.
How can I pass my Perplexity API key?
Three methods in order of priority: command‑line --api-key argument, PERPLEXITY_API_KEY environment variable, or a .env file located via the --cwd argument.
Does the server support any transport besides local stdio?
No. The README only documents configuration for local MCP client integration over stdio; no other transports are mentioned.
「AI とエージェント」の他のコンテンツ
Just Prompt - A lightweight MCP server for LLM providers
dislerjust-prompt is an MCP server that provides a unified interface to top LLM providers (OpenAI, Anthropic, Google Gemini, Groq, DeepSeek, and Ollama)
Intervals.icu MCP Server
mvilanovaModel Context Protocol (MCP) server for connecting Claude and ChatGPT with the Intervals.icu API.
Open Multi-Agent Canvas
CopilotKitThe open-source multi-agent chat interface that lets you manage multiple agents in one dynamic conversation and add MCP servers for deep research
Web Agent Protocol
OTA-Tech-AI🌐Web Agent Protocol (WAP) - Record and replay user interactions in the browser with MCP support
21st.dev Magic AI Agent
21st-devIt's like v0 but in your Cursor/WindSurf/Cline. 21st dev Magic MCP server for working with your frontend like Magic
コメント