> ## Documentation Index
> Fetch the complete documentation index at: https://docs.keenable.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Every page is listed in https://docs.keenable.ai/llms.txt. To pick a product for a job, read https://docs.keenable.ai/choose.md
> Setup steps for agents, covering the SDKs, the MCP server and the CLI: https://docs.keenable.ai/skill.md
> Keenable search and fetch work without an API key. To add them to code, use the official SDK for the project's language: https://docs.keenable.ai/sdks.md
> Without a key, call POST https://api.keenable.ai/v1/search/public or GET https://api.keenable.ai/v1/fetch/public with an X-Keenable-Title header that names your app.
> With a key, send it in the X-API-Key header and read it from the KEENABLE_API_KEY environment variable. Never ask a user to paste a key into chat. Every access method: https://docs.keenable.ai/auth.md
> To get an API key without the console, use device login: https://docs.keenable.ai/authentication.md#device-login
> MCP server: https://api.keenable.ai/mcp (no key, or X-API-Key; keyless Codex uses https://api.keenable.ai/mcp-public), or https://api.keenable.ai/mcp-oauth for OAuth sign-in. These docs are an MCP server at https://docs.keenable.ai/mcp and the OpenAPI spec is at https://docs.keenable.ai/api-reference/openapi.json
> Keenable searches and fetches public web pages. If a question needs no web information, skip the call.
> On 402 with an API key (code insufficient_credits), stop and ask a human to add credits; don't retry. On 429, wait the number of seconds in Retry-After, then retry. Before you ship, follow https://docs.keenable.ai/production.md
> Every error, with its cause and fix, is in https://docs.keenable.ai/errors.md and moving from another search API is covered in https://docs.keenable.ai/migrate.md
> Before you compare Keenable with another search API, read https://docs.keenable.ai/guides/evaluate-search-apis.md

# Rakazo web search provider

> Keenable as a built-in web provider in Rakazo, the open source self-hosted AI agent. Serves web_search and web_fetch, no key needed.

*Built-in web provider*

Give your [Rakazo](https://rakazo.com) bots web search and page reading with Keenable, a built-in web provider that needs no key. Rakazo is an open source AI agent for persistent teammates that you can self-host. Keenable serves both of Rakazo's built-in web tools: `web_search` returns titles, URLs and snippets, and `web_fetch` returns a page as readable text.

→ [View the change on GitHub](https://github.com/elie222/rakazo/pull/1522)

## Install

### No separate install

Keenable ships with Rakazo v0.3.3 and later. There is no Keenable package to install.

### Select Keenable

Rakazo's default web provider is keyless HTTP search (DuckDuckGo HTML) with local Readability extraction. To switch to Keenable, set these in the deployment's `.env`:

```bash theme={"theme":{"light":"github-light-high-contrast","dark":"dark-plus"}}
WEB_PROVIDER=keenable
# Optional; omit to run keyless.
KEENABLE_API_KEY=keen_...
```

Restart both the API and the worker, because each one builds its own web provider.

If you don't set a key, requests are anonymous and Keenable limits them per IP address. If `KEENABLE_API_KEY` is set on both the API and the worker, Rakazo sends it to Keenable and the per-IP limit no longer applies. To go back to the default, set `WEB_PROVIDER=keyless-http` or remove the line.

### Use it

Ask a bot something that needs the web. The thread shows `Searching the web: ...` while `web_search` runs and `Reading page: ...` while `web_fetch` reads a page.

## How it works

* `web_search` calls Keenable's `search_web_pages` tool on the hosted MCP server at `https://api.keenable.ai/mcp`. It returns 5 results by default and up to 10.
* `web_fetch` calls `fetch_page_content` on the same server. It returns 8,000 characters by default, up to 50,000 when the bot asks for more, and marks longer pages as truncated. Rakazo checks that the URL points to a public address before the request leaves your deployment.
* Keenable receives search queries and the page URLs a bot asks to read, plus an `X-Keenable-Title: rakazo` header that names the app. Rakazo sends no user, bot or conversation identifiers.
* Rate-limit and other provider errors go back to the bot as tool errors. Rakazo doesn't fall back to another provider.
* A bot's **Disabled tools** setting still turns either tool off.

## Related pages

> [card] [SearXNG](https://docs.keenable.ai/integrations/searxng.md)
>
> Add Keenable results to a SearXNG instance with the built-in keenable engine (shortcut !keen), enabled in settings.yml.

> [card] [PipesHub](https://docs.keenable.ai/integrations/pipeshub.md)
>
> Use Keenable as the web search provider for PipesHub assistants, from workspace settings or the admin API. No API key needed, in PipesHub 1.1.0 and later.

> [card] [Vellum Assistant](https://docs.keenable.ai/integrations/vellum-assistant.md)
>
> Keenable as a built-in web-search provider in Vellum Assistant, alongside Brave, Perplexity, Tavily and Firecrawl.

> [card] [Search quickstart](https://docs.keenable.ai/products/search.md)
>
> 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.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.