> ## 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.

# Vellum Assistant

> Keenable as a built-in web-search provider in Vellum Assistant, alongside Brave, Perplexity, Tavily and Firecrawl.

*Search provider*

Keenable as a built-in web-search provider in Vellum Assistant, alongside Brave, Perplexity, Tavily and Firecrawl.

→ [View the change on GitHub](https://github.com/vellum-ai/vellum-assistant/pull/39431)

## Install

### No separate install

Keenable ships as a built-in web-search provider in Vellum Assistant (merged upstream) — there is no separate package to install.

### Point the web-search service at Keenable

Edit `~/.vellum/workspace/config.json` (the workspace root follows `VELLUM_WORKSPACE_DIR` when that is set):

```json theme={"dark"}
{
  "services": {
    "web-search": { "provider": "keenable" }
  }
}
```

### Run it

The key is read from the environment, and the CLI forwards it into self-hosted containers and VMs.

```bash theme={"dark"}
export KEENABLE_API_KEY=keen_...
vellum wake
```

<Warning>
  **Known defect in the merged provider.** Search results currently render as a title, a URL and a published date with no text, because the provider reads a field Keenable leaves empty for most pages. The fix is open upstream as [PR #40928](https://github.com/vellum-ai/vellum-assistant/pull/40928); there is no workaround in the meantime.
</Warning>

### Note on the Settings picker

Keenable is not in the desktop/web Settings provider picker yet — that card gates BYOK providers behind a required API key field, and our provider entry does not declare one. For now select it through `config.json` or the env var.

<Note>
  Set `KEENABLE_API_KEY` to authenticate — this removes the hourly request cap. Without a key, calls run on the shared public tier at [lower rate limits](/rate-limits).
</Note>