Skip to main content
Glama
pminervini

Deep Research MCP

by pminervini
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
HF_TOKENNoHugging Face token (optional, for gated models).
GEMINI_API_KEYNoAPI key for Gemini provider. Alternatively, use GOOGLE_API_KEY. Required if using Gemini.
GOOGLE_API_KEYNoAlternative API key for Gemini provider.
OPENAI_API_KEYNoAPI key for OpenAI provider. Required if using OpenAI (default).
SERPER_API_KEYNoAPI key for Serper (optional, used for Open Deep Research).
DR_TULU_API_KEYNoAPI key for DR-Tulu provider. Required if using DR-Tulu.
OPENAI_BASE_URLNoBase URL for OpenAI-compatible endpoint.
SERPAPI_API_KEYNoAPI key for SerpAPI (optional, used for Open Deep Research).
RESEARCH_API_KEYNoAlternative API key for DR-Tulu provider.
RESEARCH_PROVIDERNoProvider to use: openai, dr-tulu, gemini, or open-deep-research. Defaults to openai if not set.

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
deep_researchA

Performs autonomous deep research using the configured provider with web search and analysis capabilities.

What it does:

  • Decomposes complex queries into research strategies

  • Conducts real-time web searches for current information

  • Executes code for data analysis and visualization (when include_analysis=True)

  • Synthesizes findings into comprehensive reports with citations

Best for:

  • Current events and recent developments

  • Data-driven analysis requiring statistics/charts

  • Complex topics needing multiple source synthesis

  • Academic-style research with proper citations

Returns: Structured markdown report with citations, metadata, and research insights.

Note: Uses provider-native research backends - monitor costs as research can generate substantial tokens.

research_statusA

Check the current status and progress of a running research task.

Use when:

  • Research is taking a long time (>5 minutes)

  • Want to check if a task completed successfully

  • Need creation/completion timestamps

Returns: Task status ('running', 'completed', 'failed') with timestamps.

research_with_contextA

Perform research using an enriched query based on clarification answers.

Use after:

  • Calling deep_research with request_clarification=True

  • Receiving clarifying questions and a session ID

  • Gathering answers from the user

What it does:

  • Takes your answers to clarifying questions

  • Creates an enriched, more specific research query

  • Performs comprehensive research with the enhanced query

Returns: Complete research report with citations and metadata

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

A3.9/5.0

Scored across 3 tools

Disambiguation5/5

Each tool has a distinct purpose: deep_research performs primary research, research_status monitors progress, and research_with_context handles enriched queries after clarification. No confusion between them.

Naming Consistency4/5

All names contain 'research', but one uses 'deep_research' while the others use 'research_', which is a minor inconsistency. Still, the pattern is mostly predictable.

Tool Count4/5

Three tools cover the core research workflow adequately for a focused server. The count is slightly low but not insufficient for the stated purpose.

Completeness3/5

The tools support initiating research, checking status, and enhanced context, but lack retrieval of past results, task cancellation, or history listing, which are notable gaps.

Maintenance

ActivitySlowing
ResponsivenessResponsive

Resources