Skip to main content
Web provider in hyperresearch Keenable is a web provider in hyperresearch, the deep-research pipeline for Claude Code and Codex. It handles both halves of the pipeline’s web work: searching, and fetching pages into the research vault. Search results already carry page text, so there is no extra fetch per result. It works without an API key and needs no extra package. → Web providers in the hyperresearch README

Install

The keenable provider ships in hyperresearch 0.13.1 and later. Install it, then add it to your coding agent from the project folder:

Select Keenable

Set the provider in the project’s .hyperresearch/config.toml:
Run a research task as usual, with /hyperresearch <question> in Claude Code or $hyperresearch <question> in Codex. Searches and page fetches now go to Keenable, and every fetched source lands in the vault under research/notes/.

Providers compared

The README lists the other providers (crawl4ai, Exa, Tavily, Parallel, Serply) and what each needs. The provider caps each request at 50 results. To lift the keyless rate limits, export a key before you run the agent:
Without a key, requests run on the shared public tier at lower rate limits, which a deep research run can reach. See API keys.

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.

Obscura

Add a web_search tool to Obscura’s MCP server with OBSCURA_WEB_SEARCH=keenable, so the agent finds pages before it browses them. No key needed.

de-anthropocentric research engine

Keenable is one of the MCP servers the de-anthropocentric research engine recommends for web retrieval, with no install and no key.

Integrations

Connect Keenable web search to your agent stack: LangChain, LlamaIndex, Haystack, MCP and more.