Skip to main content
Glama
jujubytes-a11y

Perplexity MCP Server

by jujubytes-a11y
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
MCP_HOSTNoHTTP bind host. Default: 127.0.0.1127.0.0.1
MCP_PORTNoHTTP bind port. Default: 80008000
MCP_TRANSPORTNoTransport mode: stdio or streamable-http. Default: stdiostdio
PERPLEXITY_RPMNoSoft client-side requests/minute limit. Default: 3030
PERPLEXITY_API_KEYYesPerplexity API key (required)
PERPLEXITY_MAX_RETRIESNoRetries on 429/5xx and transport errors. Default: 33
PERPLEXITY_SEARCH_TIMEOUTNoSearch timeout in seconds. Default: 6060
PERPLEXITY_MAX_CONCURRENCYNoMax concurrent API calls. Default: 22
PERPLEXITY_DEEP_RESEARCH_TIMEOUTNoDeep research timeout in seconds. Default: 180180

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

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
perplexity_searchA

General web search via Perplexity Sonar (default model: sonar-pro).

Returns JSON with answer, citations, and optional search_results/usage.

When to use optional filters:
- search_recency_filter: current events / "latest" questions (hour|day|week|month|year).
- search_after/before_date_filter: absolute windows as MM/DD/YYYY; prefer over recency when exact.
- search_domain_filter: trusted sources (allowlist) or exclude noise (denylist with '-'); one mode only.
- search_mode: 'academic' for papers, 'sec' for SEC filings; omit for general web.
- Leave filters unset for broad evergreen queries.
perplexity_deep_researchA

Comprehensive synthesis via Perplexity Sonar Deep Research.

Returns JSON with answer, citations, and optional search_results/usage.
Uses a longer timeout suitable for exhaustive research.

When to use optional filters:
- search_recency_filter: current events / "latest" questions (hour|day|week|month|year).
- search_after/before_date_filter: absolute windows as MM/DD/YYYY; prefer over recency when exact.
- search_domain_filter: trusted sources (allowlist) or exclude noise (denylist with '-'); one mode only.
- search_mode: 'academic' for papers, 'sec' for SEC filings; omit for general web.
- Leave filters unset for broad evergreen queries.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.6/5.0

Scored across 2 tools

Disambiguation3/5

Both tools target web search via Perplexity with nearly identical descriptions, differing only in timeout and 'deep research' label. An agent may struggle to choose between them without clear use-case differentiation.

Naming Consistency5/5

Both tool names follow a consistent pattern: 'perplexity_search' and 'perplexity_deep_research' use the same prefix and verb_noun structure, making them clearly identifiable.

Tool Count3/5

With only 2 tools, the set is minimal but arguably covers the server's purpose (search and deep research). However, the overlap reduces the value of having two separate tools.

Completeness2/5

The set only provides two search variants, lacking any additional functionalities like result filtering, history, or configuration. Important search UX features are missing, making the surface incomplete for a comprehensive search server.

Maintenance

ActivityStale
ResponsivenessNo issues

Resources