One coding agent. One repo. Five steps.
The agent does the work. Excellent starts it, observes it, checks it, and tests changes to the setup before anything is promoted.
1. Run the agent through Excellent.
Excellent starts the coding agent in an isolated worktree and records the exact agent version, model, instructions, tools, repo state, and task.
excellent run --agent "your-agent" --task ISSUE-4382. Check what happened.
Excellent treats done as a claim, not a result. It freezes the resulting tree and gathers evidence from the repo and tools connected to the task.
Git tree and commit Test runner Type checker Browser test Build CI API Database Deployment3. Return what is still wrong.
A returned attempt includes the unmet check, the observed result, the expected result, the supporting evidence, and the next action.
Verified Returned Needs a decision Not checked4. Compare better ways of running the agent.
Once Excellent has enough representative work, it establishes a baseline and tests candidate setups against work the candidate system did not see.
Instructions Model selection Tool descriptions Context selection Planning Check timing Retry behavior Stopping rules5. Promote only what holds up.
A candidate moves through historical comparison, held-out tasks, shadow, canary, promotion, and ongoing observation. A higher average score is not enough.
Historical comparison Held-out tasks Shadow Canary Promotion Ongoing observation
The agent does not grade itself.
The producing agent's words are useful context. They are not independent proof.
The strongest checks come from systems that can directly observe the requested result: the repo, test runner, browser, CI, API, database, or deployment.
Run the first task.
excellent init
excellent run --agent "your-agent" --task ISSUE-438