Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

TestDriver Agent Behavior Rules

Defines how the TestDriver GitHub bot reviews pull requests, builds tests from issues, and responds to @mentions.

Dev & CodingIntermediate★ 242⑂ 35AI score 6/10Last updated: Sep 10, 2026

What it does

This skill documents the operating rules for the TestDriver agent living inside a GitHub repository.

  • On PR opened: silently spins up a sandbox, runs the real app, writes and executes a TestDriver test for the change, and posts a single review comment.
  • Evidence labeling: every finding is tagged observed (proven by running) or suspected (inferred from the diff); signed-out installs get suspected only.
  • On issue opened: switches to onboarding/test-building mode and finishes by opening a PR with the test it wrote.
  • On @mention: becomes chatty and interactive with progress updates and sandbox screenshots, overriding the silent auto-review mode.
  • Scope guardrails: it only writes, debugs, fixes, and maintains tests — it declines app code edits and general coding help, and it never approves or blocks a merge.

Who it's for

  • Teams already using (or evaluating) TestDriver for computer-use E2E testing
  • QA and frontend engineers who want PR reviews backed by an actual run, not just diff reading
  • DevOps owners designing non-blocking bot review policies on GitHub

Examples

  1. A PR touches the checkout form; the agent drives the checkout flow in a sandbox and reports a broken edge case labeled observed.
  2. You open an issue asking for a "login then dashboard" test; the agent writes it and opens a PR containing the test.
  3. You @mention the bot on an existing PR to get the full play-by-play with screenshots, or send @bot logout to revoke the shared sign-in and tear down the sandbox.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/testdriverai/testdriverai/HEAD/ai/skills/testdriver-agent/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the ai/skills/testdriver-agent folder from the GitHub repo testdriverai/testdriverai into my ~/.claude/skills/testdriver-agent/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/testdriverai/testdriverai.git /tmp/testdriverai && mkdir -p ~/.claude/skills && cp -r /tmp/testdriverai/ai/skills/testdriver-agent ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repository: git clone https://github.com/testdriverai/testdriverai.git
  3. Create the skills folder: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r testdriverai/ai/skills/testdriver-agent ~/.claude/skills/
  5. Verify that ~/.claude/skills/testdriver-agent/SKILL.md exists.
  6. Restart Claude Code and ask about TestDriver agent behavior to load the skill.
  7. For real automated reviews, install the TestDriver GitHub App on your repository and sign in — without sign-in the agent can only produce suspected findings.
View source on GitHub ↗