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

# find_tools

> Discover the right tool from natural-language intent.

Semantic search across all of AgentKey's tools. Pass the user's full,
natural-language intent — the router uses embeddings plus platform-alias and
intent-keyword boosting to rank matches.

<ParamField path="q" type="string" required>
  The user's original phrasing — not a single extracted keyword. Works in
  multiple languages. Examples: `"search the latest OpenAI news"`,
  `"find trending posts on X about GPT"`, `"scrape the article at https://example.com"`.
</ParamField>

### Returns

A ranked list of tools, each with:

* **name** — the canonical `Provider/Operation` identifier
* **summary** — what the tool does
* **cost** — per-call price in credits

<Tip>
  Don't pre-extract a keyword. Passing the full intent lets both the action verb
  ("search", "scrape", "trending") and any platform mention reach the router.
</Tip>

## Example

```text theme={null}
find_tools("what people are saying about Claude on X today")
```

```json theme={null}
[
  { "name": "Twitter/searchTweets", "summary": "Search recent public tweets", "cost": { "credits_per_call": 0.5 } }
]
```

## Next step

Pass the chosen `name` to [`describe_tool`](/api-reference/describe-tool) to get
its parameters and a ready-to-run template.

<Note>
  To browse instead of search, use `list_tools(prefix)` — it walks the category
  tree (`search`, `scrape`, `social`, `crypto`, `finance`, `business`,
  `ecommerce`).
</Note>