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

# Search a Collection

> Scope a search to one collection instead of your whole library.

Pass `collection_id` to restrict a search to one collection's videos. This is faster than searching your entire account, since fewer vectors need to be compared, and it's more precise when your library holds unrelated content — a "Sales calls" collection and a "Product demos" collection shouldn't surface each other's moments.

```bash theme={null}
curl -X POST https://api.pureframe.ai/v1/search \
  -H "Authorization: Bearer pf_..." \
  -F "query=pricing objection" \
  -F "collection_id=col_abc123"
```

Text queries require `collection_id` — there's no whole-library text search. Image queries can omit it to search across every collection in your account.

## Discovering collection IDs

```bash theme={null}
curl https://api.pureframe.ai/v1/collections \
  -H "Authorization: Bearer pf_..."
```

If you're building an agent integration, `list_collections` (see [MCP Overview](/agents/mcp-overview)) does the same thing as a tool call.

## Searching across everything

Omit `collection_id` entirely on an image search to check your whole account at once — useful for something like scanning every collection for a specific logo or product appearance. See [Collections](/video-processing/collections) for how videos get organized into collections in the first place.