code-review-and-quality
Skill5 starsChanged 2 months agoApache-2.0View on GitHub
Review changed code for verified, consequential defects and applicable project-rule violations while filtering noise.
84 other projects use a version of this.
Adopted by
Most recently taken up by
- zademy/lu-querytheir own version, first seen Nov 16, 2026
- sirius-cc-wu/sirius-skillstheir own version, first seen Nov 16, 2026
- tkbk-online/skillstheir own version, first seen Nov 15, 2026
- slantview/pi-agentstheir own version, first seen Nov 15, 2026
- addyosmani/agent-skillstheir own version, first seen Nov 14, 2026
- thistleknot/skillstheir own version, first seen Nov 8, 2026
Used with
Projects with it also use constraint-driven-development, observability-and-instrumentation and 9 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
MrCipherSmithSkillconstraint-driven-development
Use when a project needs a persistent, measurable quality bar — starting a project without documented standards, 'set up constraints', or when coverage/perf/a11y thresholds get re-debated per-PR. Produces CONSTRAINTS.md with enforceable numbers, ratchets, and a guard against silent weakening. NOT for throwaway prototypes or a one-off CI setup.
Also in 16 projects that use this
HK-hubSkillobservability-and-instrumentation
Instruments code so production behavior is visible and diagnosable. Use when adding logging, metrics, tracing, or alerting. Use when shipping any feature that runs in production and you need evidence it works. Use when production issues are reported but you can't tell what happened from the available data.
Also in 35 projects that use this
HK-hubSkillcode-simplification
Simplifies code for clarity. Use when refactoring code for clarity without changing behavior. Use when code works but is harder to read, maintain, or extend than it should be. Use when reviewing code that has accumulated unnecessary complexity.
Also in 59 projects that use this
seikaikyoSkillinterview-me
需求不明時的意圖萃取訪談:一次一題、每題附上自己的猜測、聽出「真正想要 vs 覺得應該要」,直到能預測使用者反應(約 95% 信心)才動工。適用:需求缺少對象 / 動機 / 成功標準 / 約束,或使用者點名「訪談我」「先確認一下」「我們確定嗎」。明確自足的指示、純資訊查詢、機械性操作不適用。
Also in 37 projects that use this
OpenAISkillsecurity-ownership-map
Analyze git repositories to build a security ownership topology (people-to-file), compute bus factor and sensitive-code ownership, and export CSV/JSON for graph databases and visualization. Trigger only when the user explicitly wants a security-oriented ownership or bus-factor analysis grounded in git history (for example: orphaned sensitive code, security maintainers, CODEOWNERS reality checks for risk, sensitive hotspots, or ownership clusters). Do not trigger for general maintainer lists or non-security ownership questions.
Also in 3 projects that use this
JetBrainsSkillmodern-web-guidance
Search tool for modern web development best practices. MANDATORY: Execute FIRST for all HTML/CSS and clientside JS tasks. Do NOT skip — web APIs evolve rapidly and training weights contain obsolete patterns.
Also in 4 projects that use this
- +2 more from HK-hub
Other versions84 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 84
- addyosmani/agent-skills★ 118ksame name
- sanity-io/sanity★ 6.4ksame name
- ollaya-dev/ollaya★ 1.4ksame name
- SkafteNicki/dtu_mlops★ 807same name
- epam/ai-dial-chat★ 501same name
- shashankswe2020-ux/whoop-mcp★ 181same name
- muratgur/ordinus★ 111same name
- enes-alatas/spool★ 48same name
- vinvcn/addyosmani-agent-skills-zh★ 46same name
- jmxt3/gitscape.ai★ 33same name
74 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- address-review-findingsSkill
- doubt-driven-developmentSkill
- incremental-implementationSkill
- performance-optimizationSkill
- pr-descriptionSkill
- re-review-deltaSkill
- security-and-hardeningSkill
- spec-driven-developmentSkill
- test-driven-developmentSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 77 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 27
Follow the shared writing rules supplied by use_skill.
“Common” is measured: kept by at least 6 in 10 of the 77 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
High-signal code review
Issue selection adapts OpenAI's public Codex review rubric. Candidate verification follows the method described for Claude Code Review. Switchboard's system prompt defines the existing exact-head, read-only review and structured verdict contracts.
Select findings
For each possible finding, ask:
- Introduced here? Trace it to the PR diff. Do not report a pre-existing problem merely because a changed file exposed it.
- Verified? Read enough of the relevant caller, callee, test, or contract to show the failure. Name the input or state and the wrong result. For a project-rule violation, identify the exact rule that applies to this file. Discard a suspicion you cannot substantiate in the review.
- Consequential? Would the author reasonably fix this PR for the demonstrated defect, security risk, data loss, broken behavior, significant performance regression, or explicit project contract? Respect the rigor and conventions already used in this repository. Leave out cosmetic preferences, optional cleanup, speculative improvements, and issues that formatting, lint, typecheck, or CI already reports. Identify the affected caller or behavior; the possibility that something elsewhere could break is insufficient.
Before publishing a candidate, try to disprove it. Check for an existing guard, a caller that rules out the proposed input, a contract that permits the result, or an intentional behavior change described by the PR. Read enough related code to resolve the claim. Discard a candidate whose failure path or practical impact remains unsupported. A bug that needs a particular realistic input or existing state is valid when you can show that path; it need not fail for every input.
Severity describes verified impact, not confidence. Keep the canonical blocking|major|minor|nit vocabulary and the configured severity gate. A minor is a real, bounded defect worth fixing; optional polish does not become a minor because you are certain it would improve the code. The normal review omits polish suggestions instead of producing a nit list.
Switchboard's explicit spec, test-guard, and unit-contract checks remain applicable project rules. Verify their scope and the diff before filing their required findings. Do not turn a generic wish for more tests into a finding when no behavior, proof, or project rule requires them.
Report
Follow the shared writing rules supplied by use_skill. Explain the affected product task, the input or state that causes the failure, and what the user observes. Define internal terms before relying on them.
The human explanation must answer five questions: What observable result is wrong? What real action or state causes it? Why does it need a change? What fact establishes the claim, and what remains unproved? What is the smallest sufficient correction? The humanFindings tool fields carry those answers. The public comment leads with the problem, then Why it matters, Evidence, and Change. Follow the field guidance and examples in the submission tool.
Describe actions and results in words the author can read without knowing the implementation. Write "Valid review requests are rejected" instead of "The hard gate rejects the typed bind". A short sentence can still be jargon. Put exact code paths and lengthy supporting evidence in For agents; keep the real trigger, effect and evidence limits in the human explanation. For a project-rule finding, state the requirement without inventing harm. Read the human section with For agents hidden before submitting. If it needs a code glossary to make sense, rewrite it. Clear wording does not rescue a candidate that fails the finding-selection rules above.
This presentation follows Google's review guidance on explaining why and making importance clear, and Conventional Comments on a clear subject with consistent labels. The labels help the reader find the answers; they do not prove that the answers are useful or correct.
Submit one structured finding per distinct cause, with the failure scenario, impact, and a practical fix. Collapse duplicate symptoms under one finding. Do not inflate severity to force a fix round. If nothing passes the filter, approve with an empty findings list. Follow the system prompt's submit_verdict, head, severity-gate, and final-message contracts.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

