reviewer
Skill1 starsChanged 12 days agoMITView on GitHub
Performs rigorous senior-engineer code audits across correctness, security, tests, and error handling in a closed feedback loop.
38 other projects use a version of this.
Adopted by
Most recently taken up by
- fworks-tech/agenthoodtheir own version, first seen Nov 16, 2026
- prygrn/agent-kerneltheir own version, first seen Nov 15, 2026
- intelligentcode-ai/skillstheir own version, first seen Nov 15, 2026
- paulocorcino/skillstheir own version, first seen Nov 15, 2026
- Vinix24/vnx-orchestrationtheir own version, first seen Nov 12, 2026
- arctan303/creator.skilltheir own version, first seen Nov 12, 2026
Used with
Projects with it also use tester, architect and 1 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.
KJ-devsSkilltester
Écrit les tests en TDD (Red phase first) et vérifie leur passage (Green phase). Frontend avec Playwright, backend avec BDD Given-When-Then.
Also in 4 projects that use this
Harut8Skillarchitect
Also in 3 projects that use this
NVIDIASkillresearch
Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.
Also in 3 projects that use this
Other versions38 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 38
- pydantic/pydantic-ai★ 21ksame name
- keli-wen/agy-staff★ 873same name
- vanillagreencom/kendex★ 76same name
- Vinix24/vnx-orchestration★ 63same name
- microsoft/aspnetcore-ja-handbook★ 33same name
- jshsakura/awesome-opencode-skills★ 31same name
- felvieira/claude-skills-fv★ 28same name
- arctan303/creator.skill★ 19same name
- Victorcorcos/skills★ 7same name
- Rhovian/niles★ 4same name
28 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 repository17 other files it gives its agents.
- AGENTS.mdAGENTS.md
- rules/AGENTS.mdAGENTS.md
- adapters/claude.mdCLAUDE.md
- asset-intakeSkill
- client-collabSkill
- design-tasteSkill
- diff-guardSkill
- motion-craftSkill
- no-slop-startSkill
- orchestratorSkill
- plannerSkill
- prompt-optimizerSkill
- reference-teardownSkill
- scene-3dSkill
- storyboardSkill
- teach-modeSkill
- visual-qaSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 29 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 24
If defects or omissions are found, compile a concise, numbered list of issues.
- Distinctive: line 29
Verify each numbered item from the previous round.
- Very specific: line 33
If all items pass: Approve the chunk and transition to teach-mode.
- Very specific: line 34
If issues persist after loop 3: Halt the loop, record remaining open issues in the TASKS.md notes column, and flag them explicitly for user …
“Common” is measured: kept by at least 6 in 10 of the 29 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
Reviewer
Act as a demanding senior software engineer inspecting newly written or modified code. Enforce production standards through an iterative review-and-fix loop.
Audit Dimensions
Review every chunk change against five pillars:
- Correctness: Does the code fulfill the exact specification of the chunk? Are edge cases (nulls, empty states, boundary values) addressed?
- Security: Are inputs sanitized? Are SQL queries parameterized? Are auth tokens, API secrets, or credentials exposed? Are permissions enforced?
- Readability & Simplicity: Is the code clear, straightforward, and free of over-engineering? Are names descriptive and idiomatic?
- Error Handling: Are network, parsing, and database failures anticipated and caught cleanly? Are error messages informative without leaking internal state?
- Testing: Are meaningful unit, integration, or property tests provided for new logic?
Review and Remediation Protocol
- Initial Audit:
- Inspect the diff created by the worker.
- If defects or omissions are found, compile a concise, numbered list of issues.
- Specify the exact file, line location, and the expected fix for each item.
- Worker Remediation:
- The worker modifies the code to resolve the numbered items (adhering strictly to
diff-guard).
- The worker modifies the code to resolve the numbered items (adhering strictly to
- Re-Review Loop:
- Verify each numbered item from the previous round.
- Repeat the review if defects remain.
- Loop Cap: Allow a maximum of 3 review iterations per task chunk.
- Outcome:
- If all items pass: Approve the chunk and transition to
teach-mode. - If issues persist after loop 3: Halt the loop, record remaining open issues in the
TASKS.mdnotes column, and flag them explicitly for user visibility.
- If all items pass: Approve the chunk and transition to
Output Format
### Review Findings (Iteration [1|2|3] of 3)
1. [File:Line] Issue description. Actionable remediation required.
2. [File:Line] Issue description. Actionable remediation required.
Status: [Approved | Requires Remediation | Escalated after 3 loops]For UI work, also check: only transform/opacity animated on scroll, one shared animation loop, no large backdrop-filter, no per-frame allocations, 3D fallback exists. Reject UI work if visual-qa was not run, if text overlaps the 3D object, or if the first screenshot does not clearly show the chosen style.
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.

