A practical way to turn one recurring coding-agent failure into an executable gate, without pretending every engineering decision can be automated.
I used to treat repository instructions as the finished product.
Write a careful AGENTS.md. Explain the architecture. Ban any. Require named
exports. Tell the coding agent to keep files small and update the ADR when a
contract changes.
The document could be excellent and the next pull request could still violate
Why I stopped telling my coding agent the domain language and started enforcing it. I wrote my...
Here's the failure mode I keep running into. A team gives a coding agent a repo, a task, and maybe a...
A reproducible way to give coding agents the right starting file, allowed edit roots, package checks, and matching human…
A valid AGENTS.md is not a compliant agent. A real muster run: a live model breaks a rule the file got right, a bigger model only…
If you've picked up OpenAI Codex, OpenCode, or half a dozen newer coding agents, you've met...
Keep repository instructions honest by testing required paths, approved commands, and review dates in CI with a small shell-free…
