From ultrapowers
This skill should be used when receiving code review feedback, requesting review before merging, or when completing a major feature. Covers both giving and receiving reviews with technical rigor.
npx claudepluginhub jaidhyani/jai-cc-plugins --plugin ultrapowersThis skill uses the workspace's default tool permissions.
Technical evaluation, not social performance.
Verifies tests pass on completed feature branch, presents options to merge locally, create GitHub PR, keep as-is or discard; executes choice and cleans up worktree.
Guides root cause investigation for bugs, test failures, unexpected behavior, performance issues, and build failures before proposing fixes.
Writes implementation plans from specs for multi-step tasks, mapping files and breaking into TDD bite-sized steps before coding.
Technical evaluation, not social performance.
Instead: restate the technical requirement, ask clarifying questions if needed, or just fix it. Actions over words.
Push back when the suggestion breaks existing functionality, violates YAGNI (reviewer suggesting features nothing calls), is technically incorrect for this stack, or conflicts with prior architectural decisions.
How: use technical reasoning, reference working tests/code, ask specific questions. Not defensiveness.
Before implementing a reviewer's suggestion to add a "proper" feature:
grep -r "function_or_endpoint_name" src/
If nothing calls it, it's YAGNI. Ask: "Nothing calls this. Remove it, or is there usage I'm missing?"
From the user/team lead: Trusted — implement after understanding. Still ask if scope is unclear.
From external reviewers: Verify against the codebase before implementing. Check: technically correct for this stack? Breaks existing functionality? Reason for current implementation the reviewer may not know? If suggestion conflicts with prior architectural decisions, stop and escalate instead of implementing.
If any item is ambiguous, stop. Don't implement the clear ones and ask about the rest — items may be related. Clarify everything first, then implement.
Test each fix individually. Verify no regressions.
Dispatch a review agent with:
Reply to inline review comments in the comment thread, not as top-level PR comments.