Auditing Continue-built projects in CI — score posted as sticky PR comment #12309
commitshow
started this conversation in
Config
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Continue users — quick share. I've been running my Continue-built projects through commit.show, which scores AI-assisted projects on production maturity (tests, CI, observability, Lighthouse, etc.) plus a couple of pillars specific to vibe-coding workflows (
CLAUDE.mddepth,.continuerulesif present, prompt reuse, brief integrity).Local CLI
Works from any directory, no auth required for the basics.
--jsonflag emits a v1 stable schema if you want to feed it back into Continue chat.GitHub Action
The action posts a sticky comment per PR with the score and the top concerns — re-running on the same PR updates the comment instead of stacking, so the score delta after a fix is visible at a glance.
Why I'm posting
Continue's strong suit for me is the slash-command workflow. Pairing it with an audit that knows what to weigh in AI-assisted projects (vs generic "code quality" tools that penalize the wrong things) closes the loop on quality without adding much overhead.
Asks: anyone who has chained Continue with a CI quality gate before — what worked, what didn't?
Site: https://commit.show
Rubric: https://commit.show/audit
Action: https://github.com/marketplace/actions/commit-show-audit
Beta Was this translation helpful? Give feedback.
All reactions