agentleFS
IgorGanapolskyIgorGanapolsky/mac-yolo-safeguards/.agents/skills/cloudflare-kitesurf-browser/SKILL.md

cloudflare-kitesurf-browser

Skill3 starsChanged 4 days agoMITView on GitHub

Cloudflare Kitesurf via Browser Run (?browser=kitesurf) for agent screenshots and HTML extract. Fail-closed: never claim READY without CLOUDFLARE_ACCOUNT_ID + CLOUDFLARE_API_TOKEN. Fallback is fetch HTML, not a fake PNG. Trigger: Kitesurf, Browser Run, InfoQ kitesurf, lightweight agent browser. Slash: /cloudflare-kitesurf-browser.

The file

Files
SKILL.md

Cloudflare Kitesurf (Browser Run adapter)

Sources: https://blog.cloudflare.com/kitesurf/ · https://developers.cloudflare.com/browser-run/

Steal the mechanic (ephemeral WASM browser for agents, CDP-compatible, browser=kitesurf). Do not vendor Chromium. Kitesurf is beta; not open source yet.

cd ~/workspace/git/igor/mac-yolo-safeguards
node tools/cloudflare-kitesurf-browser.js --health --json
node tools/cloudflare-kitesurf-browser.js --url "https://example.com" --action html --json
node tools/cloudflare-kitesurf-browser.js --url "https://example.com" --action screenshot --output /tmp/ks.png
node tests/test-cloudflare-kitesurf-browser.js

Needs CLOUDFLARE_ACCOUNT_ID + CLOUDFLARE_API_TOKEN (Browser Rendering - Edit) for screenshot/PDF. Wrangler OAuth alone is not enough (no Browser Run scope). Without them, screenshot returns UNAVAILABLE — never a fake file. HTML may still succeed via fetch with liveClaim=false.

NEVERALWAYS
Claim READY / SUCCESS screenshot with no credsProbe health; liveClaim only with Browser Run creds
Use Kitesurf for video, WebGL, TLS bot challenges, long authRoute those to Browser Run Chromium or BrowserOS
Hero Continuity / Mac-pairHosted VPS remains the product lock

Account (Igor): wrangler whoami → 0cae7e525b9750f258704159b9bba785. Mint API token in Cloudflare dashboard with Browser Rendering edit, store via /credentials-secure-store-and-skill.

Playground: https://kitesurf.cloudflare.app/

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.