loop-verifier
Skill1 starsChanged 8 days agoApache-2.0View on GitHub
Independent checker for maker diffs in a maker/checker split. Default stance is REJECT until evidence is strong. Use when verifying loop-produced changes, after a maker agent produces a diff, or when asked to APPROVE or REJECT a branch diff against stated criteria. Do not implement fixes in this role. Caller-supplied domain rubric (when present in the prompt) is an additional input, not part of this skill.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- Rick1330/ibex-harnesstheir own version, first seen Nov 8, 2026
- SUDARSHANCHAUDHARI/MyLooptheir own version, first seen Nov 1, 2026
- cobusgreyling/loop-engineeringtheir own version, first seen Aug 13, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- cobusgreyling/loop-engineering★ 12ksame name
- Rick1330/ibex-harness★ 57same name
- SUDARSHANCHAUDHARI/MyLoop★ 2same 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 repository56 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .apm/CLAUDE.mdCLAUDE.md
- CLAUDE.mdCLAUDE.md
- .github/workflows/CLAUDE.mdCLAUDE.md
- .cursor/rules/agent-skills.mdcCursor rule
- .cursor/rules/bats.mdcCursor rule
- .cursor/rules/github-actions-workflow.mdcCursor rule
- .cursor/rules/go.mdcCursor rule
- .cursor/rules/go-test.mdcCursor rule
- .cursor/rules/instructions.mdcCursor rule
- .cursor/rules/markdown.mdcCursor rule
- .cursor/rules/shell-script.mdcCursor rule
- .cursor/rules/terraform.mdcCursor rule
- agent-skills-reviewSkill
- changelogSkill
- ci-sweeperSkill
- docs-creatorSkill
- docs-updaterSkill
- github-actions-reviewSkill
- github-actions-validationSkill
- github-issue-autofixSkill
- github-issue-triageSkill
- github-pr-bodySkill
- github-pr-reviewSkill
- github-pr-reviseSkill
- go-reviewSkill
- go-validationSkill
- instructions-reviewSkill
- markdown-validationSkill
- refactorSkill
- shell-script-reviewSkill
- shell-script-validationSkill
- tech-debtSkill
- test-gapSkill
- agent-skills-reviewSkill
- changelogSkill
- ci-sweeperSkill
- docs-creatorSkill
- docs-updaterSkill
- github-actions-reviewSkill
- github-actions-validationSkill
- github-issue-autofixSkill
- github-issue-triageSkill
- github-pr-bodySkill
- github-pr-reviewSkill
- github-pr-reviseSkill
- go-reviewSkill
- go-validationSkill
- instructions-reviewSkill
- loop-verifierSkill
- markdown-validationSkill
- refactorSkill
- shell-script-reviewSkill
- shell-script-validationSkill
- tech-debtSkill
- test-gapSkill
The file
Files
UTILITY SKILL — generic loop checker. Not a domain entry skill.
Input
- Maker diff (branch vs base) and optional maker summary
- Stated target (required): the issue, CI failure, or review comment the maker was asked to address
- Prior REJECT JSON(s) and attempt number (optional): enables REGRESSION mode on attempt 2+
- Optional domain rubric (caller-supplied markdown in the prompt)
- Optional extra context (CI log excerpt, detect facts)
- Project path allowlist / denylist when provided
Output Specification
Report per common-output-format.md. End with a single fenced JSON object (verdict APPROVE or REJECT).
Execution Scope
USE FOR:
- INITIAL scrutiny of a maker branch for factual blockers and scope violations
- REGRESSION checks that prior open rejections were actually fixed
- Confirming the diff matches the stated target (not a different problem)
DO NOT USE FOR:
- Implementing or editing product files (checker does not write fixes)
- Replacing caller-supplied domain rubric on a loop caller
- Acting as the loop entry or maker skill
- Re-running CI or linters unless the domain rubric explicitly requires it
Reference Files Guide
- common-checklist.md (always read)
- common-output-format.md (always read)
Workflow
- Load common-checklist.md. If a caller-supplied domain rubric is in the prompt, apply its REJECT conditions in addition to the generic gates: a domain rule may add REJECT conditions or stricter thresholds, but never removes or relaxes a generic gate.
- SCOPE-01: only relevant files; honor denylist and allowlist.
- INTENT-01: the change addresses the stated target.
- HONEST-01: no disabled tests, skipped assertions, or commented-out checks as the “fix.”
- SECRET-01: no credentials or sensitive values added.
- RISK-01: classify risk per common-checklist.md; for medium or high, note human review in
reason. - On attempt 2+ (prior REJECT JSON supplied), REGRESSION mode: REJECT only if a prior open rejection is still unfixed, factually wrong, or this attempt introduced a new factual error in changed files. Do not hunt for new nits.
- Emit common-output-format.md. Default
REJECTuntil SCOPE-01, INTENT-01, HONEST-01, SECRET-01, and every domain REJECT rule pass.
Error Handling
| Condition | Severity | Action |
| Cannot read the branch diff | Fatal | REJECT with reason; do not APPROVE |
| Asked to implement or edit a fix | Fatal | Do not edit; REJECT with files: []; reason: out of scope (checker role) |
| Stated target missing from the prompt | Fatal | REJECT with files: []; reason: stated target required to check INTENT-01 |
| Domain rubric missing | Info | Apply generic checklist only |
| Tests required by rubric but cannot run | Recoverable | REJECT; name the test that could not run and why in reason; do not APPROVE on faith |
Examples
- Prompt:
Verify the maker branch against the CI failure in context - Result: Markdown Evidence plus JSON
verdict; REJECT if the diff does not address the logged failure
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.