Server facts
Perplexity API Platform MCP Server
Project
- Author
- ppl-ai
- Repository
- github.com/ppl-ai/modelcontextprotocol
- License
- MIT
- Category
- Search & Web Data
Tech
- Runtime / language
- node
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 2,573
- Last commit
- Data as of
Tags
- Perplexity
- Reasoning
- Research
- Sonar
- Web Search
Overview
Perplexity API Platform MCP Server is the official integration described in the supplied README for bringing Perplexity capabilities into AI assistants. It separates direct search results from model-generated responses: one tool returns ranked Search API results with metadata, while others use Sonar models for conversational answers, detailed research, or analytical reasoning. The ask operation uses sonar-pro, research uses sonar-deep-research, and reasoning uses sonar-reasoning-pro. Research and reasoning can optionally strip thinking tags from their responses to reduce the context passed back to the client.
The main client configuration launches the @perplexity-ai/mcp-server package through npx and supplies a Perplexity API key. Examples cover command-line clients and editor or desktop integrations. A separate HTTP deployment is documented for shared or cloud use, with configurable binding address, allowed hosts, and CORS origins; it defaults to loopback rather than all interfaces. Settings also cover request timeouts, logging, and an alternate API base URL. Corporate proxy support checks a Perplexity-specific variable before standard HTTPS and HTTP proxy variables. The troubleshooting section highlights long-running research timeouts and initialization failures caused by npx installation output. This is therefore both a retrieval and model-access integration, with different tools suited to finding sources, answering a quick question, or carrying out a deeper research task.
Install & configure
Requirements
- A Perplexity API key
- npm/npx tooling for the local command
- An MCP-compatible client
- Network access to the configured Perplexity API endpoint
npx -y @perplexity-ai/mcp-serverUse the package named in the README rather than the older config_hint package. HTTP deployment is also documented. PERPLEXITY_TIMEOUT_MS adjusts timeouts, and npx -yq is suggested for clients affected by installation output.
Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.
Config file · macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Use Settings → Developer → Edit Config when available. Do not assume a Linux configuration path; check the client documentation for your platform. Remote servers use the app’s connector flow when supported, not the local stdio file shown here. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"modelcontextprotocol": {
"command": "npx",
"args": [
"-y",
"@perplexity-ai/mcp-server"
],
"env": {
"PERPLEXITY_API_KEY": "<Perplexity API key>"
}
}
}
}Config file · Project: .cursor/mcp.json
User: ~/.cursor/mcp.json
The project path is relative to the workspace root. The tilde denotes your user home directory. Use Cursor’s MCP settings to confirm which configuration and scope your installed version is reading. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"modelcontextprotocol": {
"command": "npx",
"args": [
"-y",
"@perplexity-ai/mcp-server"
],
"env": {
"PERPLEXITY_API_KEY": "<Perplexity API key>"
}
}
}
}Config file · Workspace: .vscode/mcp.json
For user scope, use the “MCP: Open User Configuration” command if available. Do not guess a user configuration path across operating systems, profiles, containers, or remote workspaces; check the client documentation. Merge this example with existing settings; do not overwrite unrelated servers.
{
"servers": {
"modelcontextprotocol": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@perplexity-ai/mcp-server"
],
"env": {
"PERPLEXITY_API_KEY": "<Perplexity API key>"
}
}
}
}Config file · macOS/Linux: ~/.codeium/windsurf/mcp_config.json
Windows: %USERPROFILE%\.codeium\windsurf\mcp_config.json
Open the MCP configuration through Windsurf’s settings when possible to confirm the active file. Paths can differ in managed or future releases; check the client documentation if this location is not present. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"modelcontextprotocol": {
"command": "npx",
"args": [
"-y",
"@perplexity-ai/mcp-server"
],
"env": {
"PERPLEXITY_API_KEY": "<Perplexity API key>"
}
}
}
}Config file · Project scope: .mcp.json at the repository root
The project file uses mcpServers. For user scope, use claude mcp add --scope user rather than assuming .mcp.json in your home directory is the user configuration. Check the client documentation for managed settings and storage details. Merge this example with existing settings; do not overwrite unrelated servers.
{
"mcpServers": {
"modelcontextprotocol": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@perplexity-ai/mcp-server"
],
"env": {
"PERPLEXITY_API_KEY": "<Perplexity API key>"
}
}
}
}Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.
Tools
Source: repository README. Tool descriptions are a documented snapshot, not a live capability check.
| Tool | What it does |
|---|---|
| perplexity_search | Return ranked web results and metadata from the Search API. |
| perplexity_ask | Answer web-aware conversational questions using sonar-pro. |
| perplexity_research | Run detailed research with sonar-deep-research. |
| perplexity_reason | Handle analytical reasoning with sonar-reasoning-pro. |
Names, permissions, and defaults can change. Check the current upstream documentation before granting access.
Use cases
- Retrieve ranked web search results with metadata.
- Answer current-information questions using sonar-pro.
- Produce detailed research through sonar-deep-research.
- Handle analytical questions using sonar-reasoning-pro.
Strengths
- Separates direct search, conversational answers, research, and reasoning.
- Supports local stdio and shared HTTP deployment.
- Provides corporate proxy settings and configurable request timeouts.
Limitations
- Requires a Perplexity API key and connectivity to the API.
- Long research requests can need higher timeout settings.
- Some strict clients need quieter npx output to initialize successfully.
Frequently asked questions
Independent listing, not an affiliation or security review. Grant the least access needed and verify the upstream instructions before running code.
