Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

[FEATURE] docs: add Desearch MCP server page to Search & Research tools

Open Beginner friendly
  • Is something wrong?
#8,364 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
2/5
Estimated time
1-3 hours
Newbie friendliness
82/100
Issue type
Documentation
Clarity
Clearly specified
Activity status
Active
Tech stack
node.js, python
Domain
documentation

Research direction

Add docs/edge/en/tools/search-research/desearch.mdx and its navigation entry in docs/docs.json under “Search & Research.” Start by reading the existing remote MCP pages in that group and use the issue’s hosted and local connection details, tool descriptions, and read-only filter example as the source. Done means the new page follows the group’s layout, explains the connection options and paid People Search tools, and appears in navigation.

Written by the indexing model from the issue text.

Description

crewai-community vendor-pitch

Feature Area (dropdown)

Documentation

Is your feature request related to a an existing bug? Please link it here.

NA

Describe the solution you'd like

Add a docs-only page for Desearch at docs/edge/en/tools/search-research/desearch.mdx, plus one navigation entry in docs/docs.json in the "Search & Research" group. The page shows how to connect the Desearch MCP server to crewAI agents with the MCP support that crewAI already has (mcps=[MCPServerHTTP(...)] and MCPServerAdapter). No new code and no new dependencies.

What Desearch provides: web search, X (Twitter) search and post data, AI search with linked sources and a summary, public page extraction and crawling, and People Search.

How to connect:

  • Hosted (Streamable HTTP): https://mcp.desearch.ai/mcp with the header Authorization: Bearer <DESEARCH_API_KEY>.
  • Local (stdio): npx -y desearch-mcp-server with the env var DESEARCH_API_KEY (Node.js 20.18.1 or newer).

The server exposes 22 tools. 20 are read-only. people-search starts a paid People Search task and people-search-control pauses, resumes or cancels one, so the page includes a create_static_tool_filter example that keeps an agent on the read-only search tools. The page follows the layout of the existing remote MCP pages in that group.

Describe alternatives you've considered

A native DesearchTool in lib/crewai-tools. A docs page over the existing MCP support is smaller, adds no code to maintain, and stays in sync with the server's tool list.

Additional context

Both examples (the mcps=[...] DSL and MCPServerAdapter) were checked against the hosted server: they connect, list the tools, and return web search results when a tool is called directly.

This issue was drafted with the help of an AI agent, so it carries the llm-generated label.

Willingness to Contribute (dropdown)

Yes, I'd be happy to submit a pull request

Dominant language
Python
Stars
60.1k
Forks
8.9k
Avg merge
1d 5h
Merged PRs (30d)
182

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from crewAIInc/crewAI

All issues in crewAIInc/crewAI

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.