ccc-makeover

$ccc-makeover — audit, fix, re-score

CC Commander · $ccc-makeover · X-Ray score, approved fixes in impact order, and a before/after report card.

$ccc-xray owns the scan, its dimensions and its scoring rubric; this skill does not redefine them. The user's selected model is preserved.

Codex: invoke $ccc-xray and ask each picker as a plain-text question.

When to use / when not to

• Use on an unfamiliar codebase, before a release, after a sprint to measure improvement, or when someone asks "how healthy is this codebase?".
• Use $ccc-xray alone for a read-only score, $ccc-code-review (or /code-review ultra) for a line-level review, $ccc-design for a visual refresh.

What happens

1. Pick the mode

An argument (xray, makeover, report-card) or a clear intent selects the mode. Otherwise ask with AskUserQuestion (header "Makeover"):

• 🟢 X-Ray scan — score the project; nothing changes
• Makeover — scan, then fix the top recommendations one at a time with your OK
• Report card — current scores, maturity level and history from earlier makeovers

2. X-Ray

Invoke the ccc-xray skill with the Skill tool. Use its scorecard as-is: per-dimension 0–100 scores, its health bands and its fix-first rule.

3. Makeover

1. Baseline. Invoke ccc-xray with the full scope (or the --category focus). Record the overall score and each dimension score.
2. Pick targets. Take the critical rows, then the needs-attention rows, ordered by $ccc-xray's fix-first rule (security first, then CI when nothing deploys, then lowest score). Keep at most --max N (default 5). With --category CAT, keep only that dimension.
3. Dry run. With --dry-run, list the targets with their recommended fix and stop.
4. Apply, one at a time. For each target ask with AskUserQuestion:
   • 🟢 Apply — run the recommended skill or targeted fix
   • Skip — move to the next target
   • Stop — end the makeover and go to step 6 Skip the question only when the user passed --auto in this request.
5. Check each fix. After every applied target, run the project's test command (npm test, pnpm test, pytest, go test ./... or cargo test, whichever the repo uses) and git status --short. On a failure, stop and ask:
   • 🟢 Fix it — make a targeted fix and re-run the tests
   • Keep the changes and stop — report the failure as open
   • Revert this target — git restore the files it touched (asks for confirmation first, since it discards those edits)
6. Re-score. Invoke ccc-xray again with the same scope. Compute the delta per dimension and overall from the two scorecards, not from memory.
7. Record. Write tasks/makeover-<YYYYMMDD>.md: baseline and final scores, maturity level before and after, each target as applied / skipped / failed, and the test commands run with their results.