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

# Browser Sandbox

> A secure browser sandbox where agents can interact with the web.

<Info>
  Browser Sandbox is now [Interact](/features/interact), which offers the same features plus it can be controlled via prompts.
</Info>

Evocrawl Browser Sandbox gives your agents a secure browser environment where agents can interact with the web. Fill out forms, click buttons, authenticate, and more.
No local setup, no Chromium installs, no driver compatibility issues. Agent browser and playwright are pre-installed.

Available via [API](/api-reference/endpoint/browser-create), [CLI](/sdks/cli#browser) (Bash / agent-browser, Python, Node), [Node SDK](/sdks/node#browser), [Python SDK](/sdks/python#browser), [Vercel AI SDK](/developer-guides/llm-sdks-and-frameworks/vercel-ai-sdk), and [MCP Server](/mcp-server).

To add browser support to an AI coding agent (Claude Code, Codex, Open Code, Cursor, etc.), install the EvoCrawl skill:

```bash theme={null}
npx -y evocrawl-cli@latest init --all --browser
```

Each session runs in an isolated, disposable or persistent sandbox that scales without managing infrastructure.

## Quick Start

Create a session, execute code, and close it:

<CodeGroup>
  ```js Node theme={null}
  // npm install @mendable/evocrawl-js
  import Evocrawl from '@mendable/evocrawl-js';

  const evocrawl = new Evocrawl({ apiKey: "fc-YOUR-API-KEY" });

  // 1. Launch a session
  const session = await evocrawl.browser();
  console.log(session.cdpUrl); // wss://cdp-proxy.evocrawl.com/cdp/...

  // 2. Execute code
  const result = await evocrawl.browserExecute(session.id, {
    code: `
      await page.goto("https://news.ycombinator.com");
      const title = await page.title();
      console.log(title);
    `,
    language: "node",
  });
  console.log(result.result); // "Hacker News"

  // 3. Close
  await evocrawl.deleteBrowser(session.id);
  ```

  ```python Python theme={null}
  # pip install evocrawl
  from evocrawl import Evocrawl

  app = Evocrawl(api_key="fc-YOUR-API-KEY", api_base="https://api.evocrawl.com")

  # 1. Launch a session
  session = app.browser()
  print(session.cdp_url)  # wss://cdp-proxy.evocrawl.com/cdp/...

  # 2. Execute code
  result = app.browser_execute(
      session.id,
      code='await page.goto("https://news.ycombinator.com")\ntitle = await page.title()\nprint(title)',
      language="python",
  )
  print(result.result)  # "Hacker News"

  # 3. Close
  app.delete_browser(session.id)
  ```

  ```bash CLI theme={null}
  # Install the EvoCrawl CLI
  npm install -g evocrawl-cli

  # Shorthand - auto-launches a session, no "execute" needed
  evocrawl browser "open https://news.ycombinator.com"
  evocrawl browser "snapshot"
  evocrawl browser "scrape"

  # Close when done
  evocrawl browser close
  ```

  ```bash cURL theme={null}
  # 1. Launch a session
  curl -X POST "https://api.evocrawl.com/v2/browser" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json"

  # 2. Execute code
  curl -X POST "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID/execute" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "code": "await page.goto(\"https://news.ycombinator.com\")\ntitle = await page.title()\nprint(title)"
    }'

  # 3. Close
  curl -X DELETE "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY"
  ```
</CodeGroup>

* **No Driver Installation** - No Chromium binary, no `playwright install`, no driver compatibility issues
* **Python, JavaScript & Bash** - Send code via API, CLI, or SDK and get results back. All three languages run remotely in the sandbox
* **agent-browser** - Pre-installed CLI with 60+ commands. AI agents write simple bash commands instead of Playwright code
* **Playwright loaded** - Playwright comes pre-installed in the sandbox. Agents can write Playwright code if they prefer.
* **CDP Access** - Connect your own Playwright instance over WebSocket when you need full control
* **Live View** - Watch sessions in real time via embeddable stream URL
* **Interactive Live View** - Let users interact with the browser directly through an embeddable interactive stream

## Launch a Session

Returns a session ID, CDP URL, and live view URL.

<CodeGroup>
  ```js Node theme={null}
  import Evocrawl from '@mendable/evocrawl-js';

  const evocrawl = new Evocrawl({ apiKey: "fc-YOUR-API-KEY" });

  const session = await evocrawl.browser({
    ttl: 120,
    activityTtl: 60,
  });

  console.log(session.id);
  console.log(session.cdpUrl);      // wss://cdp-proxy.evocrawl.com/cdp/...
  console.log(session.liveViewUrl); // https://liveview.evocrawl.com/...
  ```

  ```python Python theme={null}
  from evocrawl import Evocrawl

  app = Evocrawl(api_key="fc-YOUR-API-KEY", api_base="https://api.evocrawl.com")

  session = app.browser(
      ttl=120,
      activity_ttl=60,
  )

  print(session.id)
  print(session.cdp_url)        # wss://cdp-proxy.evocrawl.com/cdp/...
  print(session.live_view_url)  # https://liveview.evocrawl.com/...
  ```

  ```bash CLI theme={null}
  # Launch with live view and custom TTL
  evocrawl browser launch-session --stream --ttl 120 --ttl-inactivity 60

  # Launch and save session info to file
  evocrawl browser launch-session -o session.json --json
  ```

  ```bash cURL theme={null}
  curl -X POST "https://api.evocrawl.com/v2/browser" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "ttl": 120,
      "activityTtl": 60
    }'
  ```
</CodeGroup>

```json Response theme={null}
{
  "success": true,
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "cdpUrl": "wss://browser.evocrawl.com/cdp/550e8400...?token=abc123...",
  "liveViewUrl": "https://liveview.evocrawl.com/...",
  "interactiveLiveViewUrl": "https://liveview.evocrawl.com/...",
  "expiresAt": "2025-01-15T10:40:00Z"
}
```

## Execute Code

Run Python, JavaScript, or bash code in your session. Output is returned via `stdout`; for Node.js, the last expression value is also available in `result`.

<CodeGroup>
  ```js Node theme={null}
  import Evocrawl from '@mendable/evocrawl-js';

  const evocrawl = new Evocrawl({ apiKey: "fc-YOUR-API-KEY" });

  const result = await evocrawl.browserExecute("YOUR_SESSION_ID", {
    code: 'await page.goto("https://example.com"); const title = await page.title(); console.log(title);',
    language: "node",
  });

  console.log(result);
  ```

  ```python Python theme={null}
  from evocrawl import Evocrawl

  app = Evocrawl(api_key="fc-YOUR-API-KEY", api_base="https://api.evocrawl.com")

  result = app.browser_execute(
      "YOUR_SESSION_ID",
      code='await page.goto("https://example.com")\ntitle = await page.title()\nprint(title)',
      language="python",
  )

  print(result)
  ```

  ```bash CLI theme={null}
  # agent-browser commands (default - "agent-browser" is auto-prefixed)
  evocrawl browser execute "open https://example.com"
  evocrawl browser execute "snapshot"
  evocrawl browser execute "scrape"

  # Execute Playwright Python code
  evocrawl browser execute --python 'await page.goto("https://example.com")
  print(await page.title())'

  # Execute Playwright JavaScript code
  evocrawl browser execute --node 'await page.goto("https://example.com"); document.title'

  # Execute arbitrary bash in the sandbox
  evocrawl browser execute --bash 'ls /tmp'

  # Target a specific session
  evocrawl browser execute --session <id> "snapshot"
  ```

  ```bash cURL theme={null}
  # Execute Playwright Python code
  curl -X POST "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID/execute" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "code": "await page.goto(\"https://example.com\")\ntitle = await page.title()\nprint(title)",
      "language": "python"
    }'

  # Execute Playwright JavaScript code
  curl -X POST "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID/execute" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "code": "await page.goto(\"https://example.com\"); const title = await page.title(); console.log(title);",
      "language": "node"
    }'
  ```

  ```bash cURL (Bash / agent-browser) theme={null}
  # Navigate to a page
  curl -X POST "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID/execute" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{ "code": "agent-browser open https://example.com", "language": "bash" }'

  # Get accessibility snapshot with @ref IDs
  curl -X POST "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID/execute" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{ "code": "agent-browser snapshot", "language": "bash" }'

  # Interact using @ref IDs from the snapshot
  curl -X POST "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID/execute" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{ "code": "agent-browser click @e5", "language": "bash" }'

  # Extract page content as markdown
  curl -X POST "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID/execute" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{ "code": "agent-browser scrape", "language": "bash" }'
  ```
</CodeGroup>

```json Response theme={null}
{
  "success": true,
  "stdout": "",
  "result": "Example Domain",
  "stderr": "",
  "exitCode": 0,
  "killed": false
}
```

### Handling File Downloads

Files downloaded inside a session can be captured and returned as base64. Use Playwright's download API via the execute endpoint:

<CodeGroup>
  ```python Python theme={null}
  import base64

  async with page.expect_download() as download_info:
      await page.click('a#download-link')  # Click the element that triggers the download

  download = download_info.value
  path = await download.path()

  # Optionally save to a known path
  # await download.save_as('/tmp/myfile.pdf')

  # Read and output file content as base64
  with open(path, "rb") as f:
      content = base64.b64encode(f.read()).decode()
      print(content)
  ```

  ```javascript Node theme={null}
  // Get the download URL from the link element
  const href = await page.getAttribute('a#download-link', 'href');

  // Fetch the file in the browser context and convert to base64
  const b64 = await page.evaluate(async (url) => {
    const resp = await fetch(url);
    const blob = await resp.blob();
    return new Promise((resolve) => {
      const reader = new FileReader();
      reader.onloadend = () => resolve(reader.result.split(',')[1]);
      reader.readAsDataURL(blob);
    });
  }, href);

  process.stdout.write(b64);
  ```
</CodeGroup>

<Note>
  The sandbox filesystem is ephemeral — downloaded files are lost when the session ends. To persist files, read their content within the session and save it to your own storage. Persistent profiles preserve browser state (cookies, localStorage) but not files on disk.
</Note>

## agent-browser (Bash Mode)

[agent-browser](https://github.com/vercel-labs/agent-browser) is a headless browser CLI pre-installed in every sandbox. Instead of writing Playwright code, agents send simple bash commands. The CLI auto-injects `--cdp` so agent-browser connects to your active session automatically.

### Shorthand

The fastest way to use browser. Both the shorthand and `execute` send commands to agent-browser automatically. The shorthand just skips `execute` and auto-launches a session if needed:

```bash theme={null}
evocrawl browser "open https://example.com"
evocrawl browser "snapshot"
evocrawl browser "click @e5"
```

### CLI

The explicit form uses `execute`. Commands are sent to agent-browser automatically -- you don't need to type `agent-browser` or use `--bash`:

<CodeGroup>
  ```bash Navigate & Snapshot theme={null}
  evocrawl browser execute "open https://example.com"
  evocrawl browser execute "snapshot"
  ```

  ```bash Interact theme={null}
  evocrawl browser execute "click @e5"
  evocrawl browser execute "fill @e3 'search query'"
  evocrawl browser execute "scrape"
  ```
</CodeGroup>

### API & SDK

Use `language: "bash"` to run agent-browser commands via the API or SDKs:

<CodeGroup>
  ```bash cURL theme={null}
  curl -X POST "https://api.evocrawl.com/v2/browser/YOUR_SESSION_ID/execute" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "code": "agent-browser snapshot",
      "language": "bash"
    }'
  ```

  ```javascript Node theme={null}
  const result = await app.browserExecute(sessionId, {
    code: "agent-browser snapshot",
    language: "bash",
  });
  ```

  ```python Python theme={null}
  result = app.browser_execute(
      session_id,
      code="agent-browser snapshot",
      language="bash",
  )
  ```
</CodeGroup>

## Session Management

### Persistent Sessions

By default, each browser session starts with a clean slate. With `profile`, you can save and reuse browser state across sessions. This is useful for staying logged in and preserving preferences.

To save or select a profile, use the `profile` parameter when creating a session.

<CodeGroup>
  ```js Node theme={null}
  const session = await evocrawl.browser({
    ttl: 600,
    profile: {
      name: "my-profile",
      saveChanges: true,
    },
  });
  ```

  ```python Python theme={null}
  session = app.browser(
      ttl=600,
      profile={
          "name": "my-profile",
          "save_changes": True,
      },
  )
  ```

  ```bash cURL theme={null}
  curl -X POST "https://api.evocrawl.com/v2/browser" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "ttl": 600,
      "profile": {
        "name": "my-profile",
        "saveChanges": true
      }
    }'
  ```

  ```bash CLI theme={null}
  # Launch with a profile (saves changes by default)
  evocrawl browser launch-session --profile my-profile

  # Launch with a profile in read-only mode
  evocrawl browser launch-session --profile my-profile --no-save-changes

  # Shorthand: launch with profile + execute in one step
  evocrawl browser --profile my-profile "open https://example.com"
  ```
</CodeGroup>

| Parameter     | Default | Description                                                                                                                                                                |
| ------------- | ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `name`        | —       | A name for the persistent profile. Sessions with the same name share storage.                                                                                              |
| `saveChanges` | `true`  | When `true`, browser state is saved back to the profile on close. Set to `false` to load existing data without writing — useful when you need multiple concurrent readers. |

<Note>
  Only one session can save to a profile at a time. If another session is already saving, you'll get a `409` error. You can still open the same profile with `saveChanges: false`, or try again later.
</Note>

The browser session state only saves when the session is closed. So we recommend closing the browser session when you are done with it so it can be reused. Once a session is closed, its session ID is no longer valid — you cannot reuse it. Instead, create a new session with the same profile name and use the new session ID returned in the response. To save and close it:

<CodeGroup>
  ```js Node theme={null}
  import Evocrawl from '@mendable/evocrawl-js';

  const evocrawl = new Evocrawl({ apiKey: "fc-YOUR-API-KEY" });

  await evocrawl.deleteBrowser("YOUR_SESSION_ID");
  ```

  ```python Python theme={null}
  from evocrawl import Evocrawl

  app = Evocrawl(api_key="fc-YOUR-API-KEY", api_base="https://api.evocrawl.com")

  app.delete_browser("YOUR_SESSION_ID")
  ```

  ```bash CLI theme={null}
  # Close the active session
  evocrawl browser close

  # Close a specific session
  evocrawl browser close --session <id>
  ```

  ```bash cURL theme={null}
  curl -X DELETE "https://api.evocrawl.com/v2/browser" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{"id": "YOUR_SESSION_ID"}'
  ```
</CodeGroup>

### List Sessions

<CodeGroup>
  ```js Node theme={null}
  import Evocrawl from '@mendable/evocrawl-js';

  const evocrawl = new Evocrawl({ apiKey: "fc-YOUR-API-KEY" });

  const { sessions } = await evocrawl.listBrowsers();
  console.log(sessions);

  // Filter by status
  const { sessions: active } = await evocrawl.listBrowsers({ status: "active" });
  console.log(active);
  ```

  ```python Python theme={null}
  from evocrawl import Evocrawl

  app = Evocrawl(api_key="fc-YOUR-API-KEY", api_base="https://api.evocrawl.com")

  response = app.list_browsers()
  print(response.sessions)
  ```

  ```bash CLI theme={null}
  evocrawl browser list
  evocrawl browser list active
  ```

  ```bash cURL theme={null}
  curl -X GET "https://api.evocrawl.com/v2/browser" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY"

  # Filter by status
  curl -X GET "https://api.evocrawl.com/v2/browser?status=active" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY"
  ```
</CodeGroup>

```json Response theme={null}
{
  "success": true,
  "sessions": [
    {
      "id": "550e8400-e29b-41d4-a716-446655440000",
      "status": "active",
      "cdpUrl": "wss://browser.evocrawl.com/cdp/550e8400...?token=abc123...",
      "liveViewUrl": "https://liveview.evocrawl.com/...",
      "interactiveLiveViewUrl": "https://liveview.evocrawl.com/...",
      "createdAt": "2025-01-15T10:30:00Z",
      "lastActivity": "2025-01-15T10:35:00Z"
    }
  ]
}
```

### TTL Configuration

Sessions have two TTL controls:

| Parameter     | Default       | Description                            |
| ------------- | ------------- | -------------------------------------- |
| `ttl`         | 600s (10 min) | Maximum session lifetime (30-3600s)    |
| `activityTtl` | 300s (5 min)  | Auto-close after inactivity (10-3600s) |

### Close a Session

<CodeGroup>
  ```js Node theme={null}
  import Evocrawl from '@mendable/evocrawl-js';

  const evocrawl = new Evocrawl({ apiKey: "fc-YOUR-API-KEY" });

  await evocrawl.deleteBrowser("YOUR_SESSION_ID");
  ```

  ```python Python theme={null}
  from evocrawl import Evocrawl

  app = Evocrawl(api_key="fc-YOUR-API-KEY", api_base="https://api.evocrawl.com")

  app.delete_browser("YOUR_SESSION_ID")
  ```

  ```bash CLI theme={null}
  # Close the active session
  evocrawl browser close

  # Close a specific session
  evocrawl browser close --session <id>
  ```

  ```bash cURL theme={null}
  curl -X DELETE "https://api.evocrawl.com/v2/browser" \
    -H "Authorization: Bearer $EVOCRAWL_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{"id": "YOUR_SESSION_ID"}'
  ```
</CodeGroup>

## Live View

Every session returns a `liveViewUrl` in the response that you can embed to watch the browser in real time. Useful for debugging, demos, or building browser-powered UIs.

```json Response theme={null}
{
  "success": true,
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "cdpUrl": "wss://browser.evocrawl.dev/cdp/550e8400...?token=abc123...",
  "liveViewUrl": "https://liveview.evocrawl.com/...",
  "interactiveLiveViewUrl": "https://liveview.evocrawl.com/...",
  "expiresAt": "2025-01-15T10:40:00Z"
}
```

```html theme={null}
<iframe src="LIVE_VIEW_URL" width="100%" height="600" />
```

### Interactive Live View

The response also includes an `interactiveLiveViewUrl`. Unlike the standard live view which is view-only, the interactive live view allows users to click, type, and interact with the browser session directly through the embedded stream. This is useful for building user-facing browser UIs, collaborative debugging, or any scenario where the viewer needs to control the browser.

```html theme={null}
<iframe src="INTERACTIVE_LIVE_VIEW_URL" width="100%" height="600" />
```

## Connecting via CDP

Every session exposes a CDP WebSocket URL. The execute API and `--bash` flag cover most use cases, but if you need full local control you can connect directly.

<CodeGroup>
  ```typescript TypeScript theme={null}
  import Evocrawl from '@mendable/evocrawl-js';
  import { chromium } from "playwright-core";

  const evocrawl = new Evocrawl({ apiKey: "fc-YOUR-API-KEY" });
  const session = await evocrawl.browser();

  const browser = await chromium.connectOverCDP(session.cdpUrl);
  const context = browser.contexts()[0];
  const page = context.pages()[0] || (await context.newPage());

  await page.goto("https://example.com");
  console.log(await page.title());

  await browser.close();
  await evocrawl.deleteBrowser(session.id);
  ```

  ```python Python theme={null}
  from evocrawl import Evocrawl
  from playwright.sync_api import sync_playwright

  app = Evocrawl(api_key="fc-YOUR-API-KEY", api_base="https://api.evocrawl.com")
  session = app.browser()

  with sync_playwright() as p:
      browser = p.chromium.connect_over_cdp(session.cdp_url)
      context = browser.contexts[0]
      page = context.pages[0] if context.pages else context.new_page()

      page.goto("https://example.com")
      print(page.title())

      browser.close()

  app.delete_browser(session.id)
  ```

  ```bash agent-browser theme={null}
  # Use the cdpUrl from the session response
  agent-browser open https://example.com --cdp "$CDP_URL"
  agent-browser snapshot --cdp "$CDP_URL"
  ```
</CodeGroup>

## When to Use Browser

| Use Case                                             | Right Tool                             |
| ---------------------------------------------------- | -------------------------------------- |
| Extract content from a known URL                     | [Scrape](/features/scrape)             |
| Search the web and get results                       | [Search](/features/search)             |
| Navigate pagination, fill forms, click through flows | **Browser**                            |
| Multi-step workflows with interaction                | **Browser**                            |
| Parallel browsing across many sites                  | **Browser** (each session is isolated) |

## Use Cases

* **Competitive intelligence** - Browse competitor sites, navigate search forms and filters, extract pricing and features into structured data
* **Knowledge base ingestion** - Navigate help centers, docs, and support portals that require clicks, pagination, or authentication
* **Market research** - Launch parallel browser sessions to build datasets from job boards, real estate listings, or legal databases

## Pricing

Pricing is straightforward: 2 credits per browser minute. Free users get 5 hours of free usage.

## Rate limits

For the initial launch, we allow all plans up to have up to 20 concurrent browser sessions.

## API Reference

* [Create Browser Session](/api-reference/endpoint/browser-create)
* [Execute Browser Code](/api-reference/endpoint/browser-execute)
* [List Browser Sessions](/api-reference/endpoint/browser-list)
* [Delete Browser Session](/api-reference/endpoint/browser-delete)

***

Have feedback or need help? Email [hello@evocrawl.com](mailto:hello@evocrawl.com) or reach out on [Discord](https://discord.gg/evocrawl).

> Are you an AI agent that needs a EvoCrawl API key? See [evocrawl.dev/agent-onboarding/SKILL.md](https://www.evocrawl.com/agent-onboarding/SKILL.md) for automated onboarding instructions.