Trusty Squire

Trusty Squire signs up / in to websites for you so you don’t have to.

Trusty Squire is an MCP server that lets Claude Code, Codex, Cursor, OpenCode, Goose, and other coding agents create accounts on real websites and retrieve the API keys automatically — then saves each key in an encrypted, write-only vault instead of your chat, your code, or your .env. The raw provider secret never needs to enter the agent's context, so it can't be pasted into a commit or leaked in a log.

It is not a secrets manager for keys you already have, and not a browser-automation framework you script per site. Point your agent at a service — “set up Clerk and wire in the key” — and Trusty Squire opens a real browser, works through signup or sign-in one step at a time, clears the bot-detection and email-verification steps that make operator tools stall, and captures the generated key. When a real person is required for phone verification, a hard CAPTCHA, 3-D Secure, an unsupported payment, or another decision, it stops and says so rather than pretending the signup completed.

Built to be handed the keys. Provider secrets are write-only: the agent's credential tools return references and authenticated results, never stored plaintext. Backend access is a host-scoped, rate-limited, independently revocable grant, so a leaked token is killed without rotating the provider key — and you connect Google or GitHub yourself in a real browser, so the agent never types your password. Full threat model below.

One prompt

Use Trusty Squire to create a Clerk account for this app, save the generated secret key, allow api.clerk.com for server-side requests, and wire it in without putting the raw key in chat, code, or .env.

Your coding agent plans the job. Trusty Squire operates the website, stores the generated key, and can issue your backend a scoped grant. The backend calls the provider through Trusty Squire, which injects the provider key on the server side.