OpenClaw (formerly Moltbot)

Instead of using a direct Anthropic API key, you can route requests through Cloudflare AI Gateway for unified billing and analytics. Set the following secrets using Wrangler:

• CLOUDFLARE_AI_GATEWAY_API_KEY
• CF_AI_GATEWAY_ACCOUNT_ID
• CF_AI_GATEWAY_GATEWAY_ID

npx wrangler secret put CLOUDFLARE_AI_GATEWAY_API_KEY
npx wrangler secret put CF_AI_GATEWAY_ACCOUNT_ID
npx wrangler secret put CF_AI_GATEWAY_GATEWAY_ID

OpenClaw includes a Chrome DevTools Protocol (CDP) shim for headless browser automation (scraping, screenshots, etc.).

Configuration

1. Set a shared secret for authentication:

   npx wrangler secret put CDP_SECRET

2. Set your worker's public URL:

   npx wrangler secret put WORKER_URL

3. Redeploy:

   npm run deploy

Endpoints

All endpoints require authentication via the ?secret=<CDP_SECRET> query parameter.

To use Cloudflare Browser Rendering, you must set the CDP_SECRET environment variable and configure a browser profile in your openclaw.json file. The cdpUrl must point to your worker endpoint including the secret as a query parameter.

Prerequisites:

• An environment variable named CDP_SECRET.
• A configured browser.profiles section in openclaw.json.

{
  "browser": {
    "profiles": {
      "cloudflare": {
        "cdpUrl": "https://your-worker.workers.dev/cdp?secret=..."
      }
    }
  }
}

The worker automatically creates a page target upon WebSocket connection. To interact with the page, you must listen for the Target.targetCreated event to retrieve the targetId.

const WebSocket = require('ws');
const CDP_SECRET = process.env.CDP_SECRET;
const WS_URL = `wss://your-worker.workers.dev/cdp?secret=${encodeURIComponent(CDP_SECRET)}`;

const ws = new WebSocket(WS_URL);
let targetId = null;

ws.on('message', (data) => {
  const msg = JSON.parse(data.toString());
  if (msg.method === 'Target.targetCreated' && msg.params?.targetInfo?.type === 'page') {
    targetId = msg.params.targetInfo.targetId;
  }
});

For local development, create a .dev.vars file to bypass authentication and enable debugging features.

Key variables:

• DEV_MODE=true: Skips Cloudflare Access authentication and bypasses device pairing (enables allowInsecureAuth).
• DEBUG_ROUTES=true: Enables /debug/* routes.

To enable chat platform integration, set the corresponding bot tokens as secrets and redeploy your worker.

Telegram

npx wrangler secret put TELEGRAM_BOT_TOKEN
npm run deploy

Discord

npx wrangler secret put DISCORD_BOT_TOKEN
npm run deploy

Slack

npx wrangler secret put SLACK_BOT_TOKEN
npx wrangler secret put SLACK_APP_TOKEN
npm run deploy

By default, data (configs, paired devices, history) is lost when the container restarts. To enable persistence, configure R2 storage using a backup/restore approach.

1. Create R2 API Token: In the Cloudflare Dashboard, go to R2 > Overview > Manage R2 API Tokens. Create a token with Object Read & Write permissions for the moltbot-data bucket.
2. Set Secrets: Use Wrangler to set the following credentials:

• R2_ACCESS_KEY_ID
• R2_SECRET_ACCESS_KEY
• CF_ACCOUNT_ID

Operation: A cron job syncs the configuration to R2 every 5 minutes. You can also trigger a manual backup via the /_admin/ UI.

# R2 Access Key ID
npx wrangler secret put R2_ACCESS_KEY_ID

# R2 Secret Access Key
npx wrangler secret put R2_SECRET_ACCESS_KEY

# Your Cloudflare Account ID
npx wrangler secret put CF_ACCOUNT_ID

You can use the provided scripts to perform basic browser automation tasks via the command line.

Capture a single screenshot: Run the screenshot.js script with the target URL and output filename.

Capture a multi-page video: Run the video.js script with a comma-separated list of URLs and an output filename.

# Screenshot
node /path/to/skills/cloudflare-browser/scripts/screenshot.js https://example.com output.png

# Multi-page Video
node /path/to/skills/cloudflare-browser/scripts/video.js "https://site1.com,https://site2.com" output.mp4

To reduce costs, you can configure the Cloudflare Sandbox container to sleep after a period of inactivity. Without this, the container runs 24/7 and incurs full provisioned capacity costs for memory and disk.

Set the SANDBOX_SLEEP_AFTER secret to a duration (e.g., 10m, 30m, 1h).

Note: When the container sleeps, the next request will trigger a cold start, which may take 1-2 minutes. If R2 is configured, your data will persist across these restarts.