Designing codebases for AI agents
Clear boundaries, local tests, and discoverable rules reduce the cost of every change.
Clear boundaries, local tests, and discoverable rules reduce the cost of every change.
Long-running agents need explicit boundaries, checkpoints, and proof of completion.
A short task contract cuts wasted context, wasted output, and review time.
Give the agent one small job at a time, and check each result before you ask for the next one.
Use a small evaluation loop to decide which tools deserve a place in your work.
Shared repository rules, evaluation tasks, and a review policy turn individual prompting into an engineering process.
Give coding agents evidence from the repository, not a longer story in chat.