A look inside · First page
About this book
How to let an agent test your product end to end on every pull request, and how to test the agents and plugins you ship inside other people's coding agents.
Preview only. Professional includes full access.
The Trove library · programming
Agentic QA
How to let an agent test your product end to end on every pull request, and how to test the agents and plugins you ship inside other people's coding agents. Fourteen chapters in three parts: the environment, identity, state and side effects an agent needs; a runner that holds a coding agent, the hook proof, fresh installs on every OS, behavioural evals; and the two skills, the honesty rules and the guidelines document that make it stick. Every chapter says how it was built and ends with rules an agent can execute.
- 20 pages
- 20 chapters
- Updated September 10, 2026
Professional library
What’s inside
01About this bookp. 1
02Prefacep. 2
03Part I · The agent tests the productp. 3
11Part II · Testing the agent itselfp. 11
16Part III · Making it stickp. 16
About this book
How to let an agent test your product end to end on every pull request, and how to test the agents and plugins you ship inside other people's coding agents. Fourteen chapters in three parts: the environment, identity, state and side effects an agent needs; a runner that holds a coding agent, the hook proof, fresh installs on every OS, behavioural evals; and the two skills, the honesty rules and the guidelines document that make it stick. Every chapter says how it was built and ends with rules an agent can execute.
Keep exploring
Nearby on the shelf
The Complete OpenClaw Developer Guide
By Marudi
Distributed Task Reliability for AI Agents: Idempotency, Retries & Queues
Nested Sub-Agents in OrchestKit: Adopting CC 2.1.172–2.1.173
By Yonatan Gross
The Complete Guide to the Vercel AI SDK
Building Agents in Production
Claude Mods: Building Function-Hook Plugins for Claude Code
By CandleKeep