From taco
3-model code review (Claude + Codex + Gemini) with PR comments and Ready-for-Review transition
How this skill is triggered — by the user, by Claude, or both
Slash command
/taco:review-codeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
```
/review-code <feature-name>
Prerequisite: /implement <feature-name> completed
Get the full changeset from the feature branch:
cd {worktree} && git diff main...HEAD
Also gather:
Same pattern as review-plan, but with code review criteria:
Claude (always available):
docs/coding-principles.md contextCodex (if available):
call_codex with code review promptGemini (if available):
call_gemini with code review promptUse scripts/synthesize-reviews.sh to merge:
Save to .dev/specs/{feature-name}/reviews/code-review.md
If NEEDS_FIXES:
/implement again after fixesIf SHIP:
gh pr comment --body "{review summary}"
gh pr ready
scripts/notify.shnpx claudepluginhub mildsalmon/taco-harness --plugin tacoReviews code changes using parallel personas for correctness, testing, maintainability, and conditional areas like security, performance, APIs. Merges into P0-P3 severity reports for PR prep and iterative feedback.
Performs code reviews using Codex MCP with severity-grouped findings and a merge gate. Supports fast (diff-only), full (diff + lint/build), and branch review variants.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.