templates/skills/review/code/checklist/SKILL.md
Short quality gate for code review before merge
npx skillsauth add ozerohax/assistagents review-code-checklistInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
<when_to_use> <trigger>A quick and repeatable minimum check set is needed before merge</trigger> <trigger>The team wants to standardize smoke-level PR review</trigger> </when_to_use>
<input_requirements> <required>PR diff and change description</required> <required>Acceptance criteria or expected result</required> <optional>Test/linter report</optional> </input_requirements>
<core_checks> <check priority="P0">Functional correctness: the change solves the task and does not break the baseline scenario</check> <check priority="P0">Security: no obvious vulnerabilities or sensitive-data leaks</check> <check priority="P0">Tests: critical behavior is covered, or there is rationale for not adding tests</check> <check priority="P1">Maintainability: code is readable, naming is clear, and complexity is not excessive</check> <check priority="P1">Documentation and contracts: updated when external behavior changes</check> <check priority="P1">Performance: no obvious regression on hot paths</check> </core_checks>
<pass_criteria> <rule>All P0 checks are closed</rule> <rule>Remaining findings are non-blocking for release and have an owner</rule> </pass_criteria>
<do_not> <item importance="critical">Do not turn the checklist into a long all-purpose audit</item> <item importance="high">Do not mark an item as passed without a verifiable signal</item> </do_not>
<output_requirements> <requirement>Return a short status: PASS or FAIL</requirement> <requirement>For FAIL, specify exact checklist items and risk</requirement> </output_requirements>
testing
Visual regression (baseline diffs, UI stabilization)
testing
Design detailed reproducible test cases (Given/When/Then, one check per case); not smoke/regression checklists
development
Mobile testing (native/hybrid/web, gestures, offline, interruptions)
testing
Load testing (k6/JMeter scenarios, benchmarks)