What it does

Official server exposing Perplexity's search and research answers. The difference from a plain search server is that you get a synthesized answer with sources, which is faster and harder to verify. Read the citations before repeating the claim.

Install

Terminal

npx -y @perplexity-ai/mcp-server

Client configuration

Paste this into your MCP client config (Claude Desktop, Cursor, or any client that reads mcpServers), then restart the client.

mcp.json

{
  "mcpServers": {
    "perplexity-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@perplexity-ai/mcp-server"
      ]
    }
  }
}

Use it for

Getting a cited answer inside the agent loop rather than a browser tab

Skip it if

Primary-source work; it returns a summary with links, not the links themselves

When to use it

Reach for this when the task needs search inside the agent loop. Skip it if a plain shell command or a one-off script is enough — every server you add costs context and adds a failure mode.

Other TypeScript servers

Filesystem — Gives an agent scoped read, write, and search access to directories you explicitly allow
Memory — A persistent knowledge graph stored in a local JSON file, exposed through entities, relations, and observations
Sequential Thinking — Gives the model a scratchpad for structured, revisable reasoning: numbered thoughts that can branch, revise, and conclude

Perplexity MCP