Docs

Documentation

Everything you need to wire Coworkkit into your app and give the agent something to do.

Start here

Getting started

Install the SDK and wire the closed-loop token route — a working voice copilot in five steps.

Backends

Next.js is a one-line drop-in; every other JS backend mints with mintSession. Which stacks fit, and the JS-only caveat.

Connect your coding agent

Add the Coworkkit MCP server to Claude Code or Cursor: live docs with no account, then session diagnostics and config after a one-time browser sign-in.

How it works

The closed loop, end to end: what your app declares, what Coworkkit runs, and why the key never reaches the browser.

Declaring your app

Actions, surfaces & elements

The four primitives — declare what the agent can do, where the user is, what it can see and operate, and how to steer it.

Hand mode

Let the agent operate the page, not just talk about it — arming, the dual-sided gate, and what stays off-limits.

Control & confirmation

Gate an action by how much it costs to get wrong: open, a spoken confirm, or a hard on-screen click the agent can't fake.

Polish & operate

Appearance

Give the button your brand's accent and dot style with a look code — one prop, edited from the Studio, no redeploy.

The Watch panel

See what the agent sees and did — live turns, the action timeline, latency — an in-app devtool you drop in during a build.

Reference

Advanced

Multiple surfaces, the session lifecycle, credits and metering, and testing your integration — once the basics are wired.

Configuration

Every CoworkkitProvider prop and its default — FAB position and drag, the z-index escape valve, the confirmation timeout, the app-wide control baseline, and the wiring props.

Browser & framework support

Which browsers work (Chromium, Safari, Firefox), the mic + WebRTC requirement, the mobile story, and the React / Next.js versions the SDK supports.

Troubleshooting & FAQ

The agent connects but is silent, the mic is blocked, the button never appears — the usual first-run snags and their fixes.

Reading these with an AI coding agent? Point it at llms.txt — a machine-readable index, with a plain-Markdown mirror of every page under /docs-md/.