GPT Researcher MCP Server

MCP server wrapper around GPT Researcher that exposes deep research, quick search, report writing, and source retrieval as MCP tools. Supports STDIO, SSE, and Streamable HTTP transports for integration with Claude Desktop and other MCP clients.

Evaluated Mar 06, 2026 (0d ago) vunknown

Homepage ↗ Repo ↗ AI & Machine Learning deep-research mcp autonomous-agent tavily python claude-desktop sse stdio

⚙ Agent Friendliness

/ 100

Can an agent use this?

🔒 Security

/ 100

Is it safe for agents?

⚡ Reliability

/ 100

Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality

Documentation

Error Messages

Auth Simplicity

Rate Limits

🔒 Security

TLS Enforcement

Auth Strength

Scope Granularity

Dep. Hygiene

Secret Handling

GPT Researcher MCP wrapper. Requires LLM + search API keys. Web research may surface untrusted content — validate before acting on findings.

⚡ Reliability

Uptime/SLA

Version Stability

Breaking Changes

Error Recovery

Best When

You want deep, multi-source research accessible as MCP tools from Claude Desktop or other MCP-compatible clients, and can tolerate 30-40 second response times.

Avoid When

You need instant search results, don't use MCP-compatible clients, or want to avoid managing API keys for OpenAI and Tavily.

Use Cases

• Deep research from within Claude Desktop or other MCP clients
• Automated report generation with cited sources via MCP protocol
• Investment and market research integrated into AI workflows
• n8n workflow automation with deep research capabilities
• Source validation and fact-checking within agent pipelines

Not For

• Sub-second search results - deep research takes ~30-40 seconds
• Environments without Python 3.11+
• Use cases that don't need MCP protocol (use gpt-researcher directly)
• Offline or air-gapped environments

Interface

REST API

GraphQL

gRPC

MCP Server

Yes ↗

SDK

Webhooks

Authentication

Methods: api_key

OAuth: No Scopes: No

Requires OPENAI_API_KEY and TAVILY_API_KEY passed explicitly in MCP config (not from shell env). Supports alternative GPTR-compatible retrievers (Bing, Google).

Pricing

Model: open_source

Free tier: Yes

Requires CC: No

MIT licensed. Operational costs depend on OpenAI and Tavily API usage.

Agent Metadata

Pagination

none

Idempotent

Retry Guidance

Not documented

Known Gotchas

⚠ API keys must be passed explicitly in Claude Desktop config, not from shell env or .env
⚠ Deep research takes 30-40 seconds per query
⚠ Python 3.11+ required
⚠ 323 stars - relatively small community compared to parent project
⚠ Docker recommended for production to avoid dependency issues

Alternatives

{'name': 'gpt-researcher', 'note': "The underlying library; use directly if you don't need MCP"} {'name': 'Tavily MCP', 'note': 'Simpler web search MCP without autonomous research orchestration'} {'name': 'Exa MCP', 'note': 'Neural search MCP server, faster but less depth'}

Full Evaluation Report

Detailed scoring breakdown, competitive positioning, security analysis, and improvement recommendations for GPT Researcher MCP Server.

$99

API endpoint ↗ Agent guide ↗ Report inaccuracy

Scores are editorial opinions as of 2026-03-06.