One result vocabulary
Every run ends verified, returned or unchecked, with the same fields behind each word, so a review conversation starts from the same facts.
Excellent gives every agent run the same shape: attempt, checks, evidence, result. Reviewers compare records instead of arguing from transcripts.
curl -fsSL https://excellent.so/install.sh | shOne engineer trusts a passing unit test, another wants browser evidence, and nobody can say what the agent actually observed.
Agent runs happen in local terminals, pull requests and CI logs, and the evidence is scattered across all three.
A regression gets through because the reviewer had to reconstruct the context after the fact.
Every run ends verified, returned or unchecked, with the same fields behind each word, so a review conversation starts from the same facts.
Repo state, command output, test results and reviewer notes belong to the run that produced them, not to whoever remembers them.
The worker's completion claim is kept separate from the evidence that decides whether the work moves forward.
What a check is, and what each result state commits to.
The record shape a reviewer reads, including retries.
What running Excellent across a team looks like today, and what it costs.
Where attempts, artifacts and receipts are written, and what leaves the machine.
Run one real task and see what held up
curl -fsSL https://excellent.so/install.sh | sh