Guides
Practical guides for engineering teams that let AI coding agents work on real repositories — how to keep a person in control, how to check the result, and how to review it.
-
How to Add Human Approval to AI Coding Agents
Four places to put a human checkpoint in an AI coding workflow — before the plan, before execution, before merge, and before deploy — and how to set each one up on GitHub.
-
How to Validate AI-Generated Code Before Merging
A practical setup for testing AI-written changes before they reach your main branch: your own CI commands, required status checks, an isolated environment, and a hard stop on failure.
-
From GitHub Issue to Pull Request with AI Coding Agents
How to run AI coding agents on GitHub issues without losing control: writing issues an agent can finish, triggering work with labels, reviewing the plan, and reviewing the pull request.
-
A Code Review Checklist for AI-Generated Pull Requests
What to check when reviewing a pull request written by an AI coding agent: scope, tests, dependencies, security, and the mistakes agents make that humans rarely do.
Follow new guides with the RSS feed.
Team plan — $119/month, 5 seats included. AI model usage included.