Skip to main content
Retriever plugin Give GPT Researcher Keenable web search with the gpt-researcher-keenable retriever plugin. It works without an API key. GPT Researcher finds the plugin through its plugin entry point, so you don’t change any GPT Researcher code. → gpt-researcher-keenable on PyPI

Install

You need GPT Researcher 0.16.1 or later. Install the plugin:

Select Keenable

Set the RETRIEVER environment variable:
To use Keenable alongside the built-in retrievers, separate the names with commas, for example RETRIEVER=keenable,arxiv.

Retriever behavior

  • Each result gives GPT Researcher a title, a URL and a short preview of the page. GPT Researcher then scrapes the pages it keeps, as it does with its built-in web retrievers.
  • MAX_SEARCH_RESULTS_PER_QUERY sets the number of results, from 1 to 50.
  • If you set query_domains, the plugin searches only the first domain you list.
  • If a network or API call fails, the plugin logs the error and returns no results, so the research run continues.
If you don’t set a key, searches run on the shared public tier at lower rate limits. Set KEENABLE_API_KEY to use your organization’s limits and remove the hourly cap.

Hermes Agent

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

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.

ketch

Use Keenable as the built-in search backend in ketch, the agent CLI written in Go.

Fetch quickstart

Turn any URL into clean markdown your model can read, or get only the part a prompt asks for. Try it with no API key. Code, cost per call and gotchas.