Is your codebase ready for a good AI harness?
This article was written in French and translated automatically by an LLM. Lire l'original
Your codebase might not be ready at all for a good AI harness.
When people talk about AI harnesses, everyone immediately thinks of AGENTS.md files, skills, lint rules, type checks…
But once you start doing serious agentic development, there are two other points that are absolutely necessary for the AI to verify its work in good conditions.
1. Start everything locally, easily
It must be able to start everything that makes up the program locally, easily. For example with an init.sh that takes care of everything.
Note that if you work with agents in parallel through worktrees on the same machine, you will have to isolate everything per instance (dynamic ports, database, volumes, Docker Compose project names…) to be able to run the full program several times in parallel. And plan for RAM…
It is also important that the program is in a clean state on every launch. Goodbye to the local database that lives on for months and months: every startup must get its own fresh database with seeds (and potentially different sets depending on the tests).
2. Full observability per instance
Every full instance of the program must be as observable as possible (logs, network, OpenTelemetry, metrics…) and every agent must have full access to the observability of its instance.
The result
Thanks to this development environment, agents can verify their work at runtime. And the difference in results is striking!
New instructions also become possible: “no errors in the logs”, “action under XXX ms”, etc.
Keep it in mind on your projects ;)