Skip to main content
Recommended MCP server Give your de-anthropocentric research engine (DARE) runs Keenable web search and page fetch. DARE is a library of agent skills that binds to no retrieval tool: a run searches with whatever MCP servers your coding agent already has. Its README lists Keenable among the recommended servers because it needs no install and no key. → View the change on GitHub

Install the engine

Run this in your own project directory, not in a clone of the DARE repository:

Add Keenable to your agent

DARE ships no MCP config file. Add Keenable to the coding agent that runs DARE instead. In Claude Code, put this in .mcp.json at the root of your project:
For other agents, see MCP server.

Use it in a research run

Start a run with /dare and give it a research direction. When a step needs the web, the agent can call search_web_pages and fetch_page_content. See the MCP server reference for the full tool list.
This entry sends no key, so calls run on the shared public tier at lower rate limits. To remove the hourly request cap, add "headers": { "X-API-Key": "keen_<your_key>" } to the keenable entry in .mcp.json.

GPT Researcher

Give GPT Researcher web search with no API key through gpt-researcher-keenable, a retriever plugin.

Qwen-MM-Plugins

Use Keenable for web_search and web_extractor in the Qwen-MM-Plugins search plugin with QWEN_MM_SEARCH_BACKEND=keenable. Works without an API key.

Hermes Agent

Keenable is built into Hermes Agent as a web backend for web_search and web_extract, and works without a key.

Search quickstart

Give your agent ranked web pages with excerpts ready to cite. Try it with no API key. Code, cost per mode, filters by site, date and domain, and gotchas.