Files
zeekay 600a2d1feb perf(ci): review the diff beside the tests
A stalled git.hanzo.ai gives up in twenty seconds and the run
continues on github.com. Publishing waits for the review. The
tests do not.

Testing: bash bin/get_test.sh && bash .hanzo/review/review_test.sh
Risk: high. A refused diff publishes nothing. A reviewer that cannot be reached still publishes.
How to revert: git revert this commit and move the v2 tag back to 5262d63804.
2026-10-16 23:03:30 -07:00

1.0 KiB

Review

Failure here is a green run that did not test, or a red run that blocked a change nobody read. Style is not a finding.

  • Cap nits at zero. Do not post style, diagnostics, or cleanup.
  • A finding names a file and a line in the diff. A guess from a name is not a finding.
  • Tag every finding Important, Nit, or Pre-existing. A bug this change did not introduce is Pre-existing.
  • The summary starts with a count and says when there is nothing to fix.
  • Re-review only the new diff. Do not repeat a finding already posted. After three rounds, stop and reconsider the approach.
  • An agent's approval is not a person's. The person who prompted the change is the author.
  • Risk comes from the diff.
    • Low: docs and copy. No further person once defect review and security review are green.
    • Medium: ordinary changes. One person, the author. The commit messages carry Testing, Risk, and How to revert.
    • High: iam, kms, auth, authz, billing, pay, release, installer, and any workflow file. Those three lines are required before release.