review-implementation
Skill3 starsChanged 8 days agoMITView on GitHub
Read-only evaluator for one bounded implementation target, by default an exact diff, supplied with its brief. Return causally qualified candidate findings; direct review requests belong to review-change, and repair decisions stay with the calling implementing agent.
10 other projects use a version of this.
Adopted by
Most recently taken up by
- grove/promise-to-prooftheir own version, first seen Oct 12, 2026
- Svampen/claude-skillstheir own version, first seen Oct 11, 2026
- DasDigitaleMomentum/opencode-processing-skillstheir own version, first seen Oct 9, 2026
- karrioapi/karriotheir own version, first seen Oct 8, 2026
- gtrabanco/agentic-workflowtheir own version, first seen Sep 22, 2026
- ayoubben18/ab-methodtheir own version, first seen Sep 18, 2026
Other versions10 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 10
- karrioapi/karrio★ 799same name
- ayoubben18/ab-method★ 192same name
- withzombies/hyperpowers★ 80same name
- DasDigitaleMomentum/opencode-processing-skills★ 60same name
- GiggleLiu/cryochamber★ 34same name
- gtrabanco/agentic-workflow★ 23same name
- grove/promise-to-proof★ 4same name
- julienlegoux/skills★ 4same name
- rakaadi/agent-kit★ 1same name
- Svampen/claude-skills★ 0same name
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 repository34 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/AGENTS.mdAGENTS.md
- analyze-projectSkill
- api-contract-strategySkill
- architecture-patternsSkill
- close-changeSkill
- design-changeSkill
- development-standardsSkill
- docker-multiarch-buildSkill
- error-patternsSkill
- go-guidelinesSkill
- herdr-implementSkill
- implement-changeSkill
- logging-standardsSkill
- lua-guidelinesSkill
- organize-docsSkill
- output-stylesSkill
- plan-changeSkill
- powershell-guidelinesSkill
- python-guidelinesSkill
- review-changeSkill
- review-designSkill
- review-planSkill
- routeros-scripting-guidelinesSkill
- runtime-diagnosisSkill
- security-guardrailsSkill
- shell-guidelinesSkill
- skill-authoringSkill
- skill-minerSkill
- skills-routingSkill
- smart-commitSkill
- sops-age-guardrailsSkill
- sync-truthSkill
- testing-strategySkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 10 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
When the selected mechanism produces a growing repair stream, the finding compares simplifying or removing it with further hardening.
“Common” is measured: kept by at least 6 in 10 of the 10 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
Review Implementation
Review only the supplied objective, non-goals, acceptance criteria, exact changed files or diff, declared verification, and a small justified supporting-file set, under the finding rules, causal scope, and verdict in review-change.
Implementation-Specific Checks
- Evidence applies to the actual candidate and protected behavior, not merely to a successful process, report, or source export. On follow-up, inspect the supplied new content, prior dispositions, unresolved findings, and affected regression boundaries; a repair claim needs evidence, and a changed round alone does not repeat the full audit.
- A blocking candidate is caused or newly activated by the current diff, violates a named requirement or oracle, has a concrete consequence, carries sufficient evidence, and admits a smallest fix inside the approved scope. When the brief supplies a whole artifact instead of a diff, defects anywhere in it are in scope.
- A new issue enters scope when evidence shows it blocks the current outcome or its necessary correctness or safety within the chosen design and authority; a directly blocking caller or schema mismatch and a change-induced regression qualify, while thematic relevance, a red test, an accepted label, or a small diff alone do not. A demonstrated critical risk outside authority is returned as
needs-decision. - When the selected mechanism produces a growing repair stream, the finding compares simplifying or removing it with further hardening.
- Newly introduced or changed identifiers are checked for leaked plan task IDs, stage codes, or review rounds under the naming rule in
development-standards, with evidence that the name depends on the plan rather than domain responsibility; genuine domain terminology and compatibility-bound names stand. - An authorized best-effort secondary adaptation with evidence is not a defect; relabeling a required result as secondary, deleting its oracle, or replacing a specified library without substitution authority is.
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.

