perplexity-mcp-server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LOG_DIR | No | Directory for file-based logging (only used if LOG_TO_STDERR is false) | ./logs |
| MAX_HISTORY | No | Maximum number of messages per session | 20 |
| LOG_TO_STDERR | No | Log to stderr (recommended for npx/MCP usage) | true |
| DISABLE_LOGGING | No | Disable all logging | false |
| SESSION_TIMEOUT | No | Session timeout in seconds (default 30 minutes) | 1800 |
| PERPLEXITY_MODEL | No | Model to use: sonar, sonar-pro, sonar-reasoning | sonar |
| PERPLEXITY_API_KEY | Yes | Your Perplexity API key (required) | |
| ENABLE_CONVERSATIONS | No | Enable session management for multi-turn conversations | true |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| queryA | Query Perplexity AI with a prompt. Specialized for web search with citations, sources, and related questions. Supports multi-turn conversations when sessionId is provided. Returns rich metadata (citations, sources, images, related questions). |
| searchB | Search for information using Perplexity. Returns a list of relevant search results. Follows OpenAI MCP specification for search tools. |
| fetchA | Fetch the full contents of a search result document by its ID. Follows OpenAI MCP specification for fetch tools. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
search and query both perform searches, which could cause confusion, but query is clearly described as more advanced with citations, sources, and multi-turn support, while search is more basic and spec-compliant. fetch is distinct as it retrieves full document contents by ID. The descriptions help differentiate, though some overlap remains.
All tool names are single lowercase verbs (search, query, fetch), which is consistent and predictable. There is no verb_noun pattern, but the naming convention is uniform and easy to follow.
Three tools is a minimal but appropriate size for a search-focused server. Each tool serves a clear purpose, and the count is not excessive or overly thin for the apparent scope.
The tools cover the core search workflow: basic search, advanced querying with rich metadata and multi-turn support, and fetching full contents. Minor gaps might include a dedicated tool for managing conversation history, but the surface is largely complete for common use cases.