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

# Use Cloudflare in Deepline

> Use cloudflare_crawl to crawl websites and extract content as markdown, HTML, or JSON. Includes inputs, outputs, SDK V2 examples, and Deepline cost.

[Human guide: work with your agent →](https://deepline.com/docs/human/providers/cloudflare/guide.md)
Inspect one Cloudflare action before starting provider-backed work. Your first result is a verified input and output contract.

## Provider-specific requirements

### When to use

* You need to crawl a website and extract structured content (markdown, HTML, or JSON).
* You need browser-rendered pages (JavaScript-heavy SPAs, dynamic content).
* You want a managed crawl that follows links up to a configurable depth/limit.

### Key parameters

* **`url`** (required): The starting URL.
* **`limit`**: Max pages to crawl. Default 10. Keep low unless the user needs broad coverage.
* **`depth`**: Max link depth from the starting URL. Default 100,000.
* **`source`**: URL discovery method — `"all"` (default), `"sitemaps"`, or `"links"`.
* **`formats`**: Array of `"html"`, `"markdown"`, `"json"`. Default `["markdown"]`. Markdown is best for LLM consumption. JSON uses Workers AI for extraction and requires `jsonOptions`.
* **`render`**: Browser rendering toggle. Default `true`. Set `false` for simple static pages to save browser seconds.
* **`options.includePatterns`** / **`excludePatterns`**: Wildcard patterns to filter which URLs get crawled. Exclude takes priority.
* **`timeoutMs`**: How long to poll before returning partial results. Default 5 minutes.

### Timeout behavior

* On timeout, the action returns whatever partial results are available with `timedOut: true`.
* The `jobId` is included so you can construct a follow-up poll if needed.
* The action does NOT throw on timeout.

### Job statuses

* `running` — crawl in progress (non-terminal)
* `completed` — all pages crawled successfully
* `errored` — unrecoverable error
* `cancelled_due_to_timeout` — exceeded 7-day max runtime
* `cancelled_due_to_limits` — hit account browser time limits
* `cancelled_by_user` — manually cancelled

### Cost awareness

* `render: true` crawls are billed on browser time, post-deduct, based on `browserSecondsUsed` in the response.
* `render: false` crawls use 0 browser-seconds (free during beta).
* Deepline credit pricing for these actions is generated from the provider pricing metadata and rendered on the public provider pages.
* Keep `limit` reasonable — large crawls consume significant browser time.
* Crawl jobs expire after 7 days. results available 14 days post-completion.

### Robots.txt

* Cloudflare respects `robots.txt` by default, including `crawl-delay`.
* Blocked URLs appear in results with `"status": "disallowed"`.

## Find an applicable action

Confirm your effective workspace, then search the live catalog:

```bash theme={"languages":{"custom":["/languages/deepline-prompt.json"]}}
deepline auth status --json
deepline search "Cloudflare" --type tools --json
```

Search returns candidates. Check the returned provider and operation rather than selecting a similar title automatically.

## Inspect the selected contract

Replace `<returned-tool-id>` with the identifier returned by discovery:

```bash theme={"languages":{"custom":["/languages/deepline-prompt.json"]}}
deepline describe <returned-tool-id> --json
```

These commands do not execute the selected provider action. Read its required inputs, supported access mode, pricing unit, outputs, and limits.

A connection does not establish that every action supports BYOK. Follow the provider-specific requirements above and the live action contract.

## Prepare one reviewed input

Use one authorized record that matches the inspected schema. Preserve its source identity and define the result you will accept.

Open the action reference for its payload and response shape. Provider-backed execution can incur usage or external effects.

For reusable logic, follow [Write Plays](https://deepline.com/docs/sdk-v2/plays.md). For direct-call recovery, use [Tool execution recovery](https://deepline.com/docs/sdk-v2/tool-execution-recovery.md).

> [!TIP]
>
> Continue only when the workspace, provider, required fields, acceptance rule, and applicable price are understood. Keep unavailable access separate from no-result evidence.


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