Agent Skills

Give your AI coding agent the ability to run human QA tests — without leaving your editor.

What are Agent Skills?

Agent Skills is an open standard for giving AI coding agents new capabilities. A skill is a package of instructions that teaches your agent how to use a tool — in this case, Runhuman.

When you install the Runhuman skill, your agent learns the full Runhuman CLI: how to create tests, read results, manage projects, and more. Instead of copying commands from docs, you just tell your agent what to test in plain English.

Agent Skills are supported by 30+ coding agents including Claude Code, Cursor, GitHub Copilot, Codex, Windsurf, Cline, and many others. Install once, and it works across any compatible agent.

How it works

  1. Install the skill — one command adds Runhuman knowledge to your agent
  2. Ask your agent to test — describe what to test in natural language (e.g., “test the checkout flow on staging”)
  3. A human tester runs the test — a professional tester follows your instructions on a real browser or device
  4. Results come back to your agent — structured feedback with bugs, screenshots, and observations
  5. Your agent acts on the results — it can read findings, fix issues, and re-test

The entire loop happens inside your editor. No context switching, no dashboard tabs, no manual CLI work.

Quick start

Copy and paste this into your agent to get started:

On first-time setup, your agent will open your browser to authenticate the Runhuman CLI.

What your agent can do

  • Create and run QA test jobs — describe what to test in natural language
  • Read previous job results — review and interpret test findings
  • Manage projects and templates — organize your testing workflow
  • Set up recurring schedules — automate daily, weekly, or monthly test runs from templates
  • Manage your organization — handle API keys, members, and billing
  • Set up CI integration — help you add the GitHub Action to your repo

Alternative: Need a persistent MCP connection? See the MCP guide.

Want automatic QA after every task? See Agent Workflow Integration to configure your agent to run Runhuman tests automatically as part of your development flow.