# Browser Markdown Scrape (/docs/experiments/browser-markdown-scrape)



Convert a URL to Markdown and scrape CSS selectors using Cloudflare Browser Rendering **quickAction** (REST-style binding — not Puppeteer).

## Features [#features]

* `GET /markdown?url=` → page Markdown via `quickAction("markdown")`
* `GET /scrape?url=&selector=` → structured elements via `quickAction("scrape")`
* HTTP/HTTPS URL validation only

## API Reference [#api-reference]

### GET /markdown [#get-markdown]

**`url`** `string` (required)

Target page (`http://` or `https://`).

#### Example Request [#example-request]

```bash
curl "https://your-worker.workers.dev/markdown?url=https://example.com"
```

#### Success Response [#success-response]

```json
{
  "url": "https://example.com/",
  "markdown": "# Example Domain\n\n..."
}
```

#### Error Codes [#error-codes]

* `400` - Missing or non-http(s) url (`INVALID_URL`)
* `502` - quickAction failure (`MARKDOWN_ERROR`)

### GET /scrape [#get-scrape]

**`url`** `string` (required)

Target page.

**`selector`** `string` (required)

CSS selector (sent as one scrape element).

#### Example Request [#example-request-1]

```bash
curl "https://your-worker.workers.dev/scrape?url=https://example.com&selector=h1"
```

#### Success Response [#success-response-1]

```json
{
  "url": "https://example.com/",
  "results": [
    {
      "selector": "h1",
      "results": [{ "text": "Example Domain", "html": "Example Domain" }]
    }
  ]
}
```

#### Error Codes [#error-codes-1]

* `400` - Invalid url (`INVALID_URL`)
* `400` - Missing selector (`MISSING_SELECTOR`)
* `502` - quickAction failure (`SCRAPE_ERROR`)

## Use Cases [#use-cases]

* Normalize pages to Markdown for RAG / LLM pipelines
* Extract specific DOM fields without writing Puppeteer scripts
* Compare quickAction REST bindings vs Puppeteer-based experiments

## Limitations [#limitations]

* Requires Browser Rendering (often `wrangler dev --remote` locally)
* Single CSS selector per scrape request
* JS-heavy SPAs may need advanced goto options (not exposed in this demo)

## Deployment [#deployment]

<Steps>
  <Step>
    ### Click the deploy button [#click-the-deploy-button]

    [![Deploy to Cloudflare Workers](https://deploy.workers.cloudflare.com/button)](https://deploy.workers.cloudflare.com/?url=https://github.com/shrinathsnayak/cloudflare-experiments/tree/main/apps/experiments/browser-markdown-scrape)
  </Step>

  <Step>
    ### Confirm browser binding [#confirm-browser-binding]

    `wrangler.json` declares `"browser": { "binding": "BROWSER" }`.
  </Step>

  <Step>
    ### Test your deployment [#test-your-deployment]

    ```bash
    curl "https://your-worker.workers.dev/markdown?url=https://example.com"
    ```
  </Step>
</Steps>

## Local Development [#local-development]

```bash
cd apps/experiments/browser-markdown-scrape
npm install
npm run dev -- --remote
```

```bash
curl "http://localhost:8787/markdown?url=https://example.com"
```

## Configuration [#configuration]

* **Browser binding** `BROWSER` — uses `quickAction("markdown" | "scrape", …)`

## Cloudflare Features Used [#cloudflare-features-used]

* **[Workers](https://developers.cloudflare.com/workers/)** - Edge compute runtime
* **[Browser Rendering](https://developers.cloudflare.com/browser-rendering/)** - Headless browser at the edge
* **[Markdown endpoint](https://developers.cloudflare.com/browser-rendering/rest-api/markdown-endpoint/)** - URL → Markdown quick action
* **[Scrape endpoint](https://developers.cloudflare.com/browser-rendering/rest-api/scrape-endpoint/)** - CSS selector scrape quick action