Repo profile governance
Skill chf3198/megingjord-harness/skills/repo-profile-governance
Audit and harden repository profile, community health, discoverability metadata, and contribution surfaces across repos using bounded, evidence-based checks.From its SKILL.md
npx -y skills add chf3198/megingjord-harness --skill repo-profile-governanceAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
4.5 KB, 927 tokens by cl100k_base, as published. Nobody here has run it
Repo Profile Governance
Purpose
Provide a bounded governance pass for repository presentation and collaboration readiness:
- profile quality (
description,homepage, topics, social preview) - community health files (
CONTRIBUTING,CODE_OF_CONDUCT,SECURITY,SUPPORT) - contribution surfaces (issue/PR templates, config, ownership cues)
- consistency across repositories
Hard constraints
- No unbounded loops or recursive retries.
- Maximum one full repository pass per invocation.
- Maximum ten recommendations per invocation.
- No silent policy mutation; produce explicit, auditable proposals.
- No claims of completion without verification evidence.
- If required artifacts are missing, return
NO_CHANGEwith missing evidence.
Modes
profile-audit: detect profile/community gaps and rank by severity.profile-remediate: propose minimal, ordered remediation steps for top gaps.profile-weekly-check: compact drift check for recurring hygiene regressions.
Scope boundary (primary ownership)
This skill is the primary owner for repository profile hygiene:
- About metadata quality (
description, homepage, topics) - discoverability surfaces (social preview, naming clarity)
- contributor-facing files and templates
- community-health baseline and consistency across repos
This skill is not the primary owner for branch protection/rulesets, merge gates,
release readiness, or incident execution flow. Delegate those to github-ops-excellence.
Required checks
A) Repository profile & discoverability
- About
descriptionis concise and accurate. homepageis set when applicable.- Topics are present, normalized, and relevant.
- Social preview image exists and is intentional (not fallback-only).
B) Community health baseline
- Presence of
README.mdwith clear purpose and quick-start/usage context. - Presence (repo root or
.github/) of:CONTRIBUTING.mdCODE_OF_CONDUCT.mdSECURITY.mdSUPPORT.md
LICENSEexists when distribution requires it.
C) Contribution surfaces
- Issue templates exist and are aligned to common workflows.
- PR template exists and enforces evidence/checklist quality.
.github/CODEOWNERSexists for critical paths.
D) Governance consistency
- Critical standards are consistent across target repos.
- Deviations are documented and intentional.
- Missing items are prioritized by impact:
- P1: security/reporting or contributor-blocking
- P2: discoverability/collaboration quality
- P3: polish/consistency
Output format (required)
PROFILE_GOVERNANCE_REPORT
mode: <profile-audit|profile-remediate|profile-weekly-check>
scope: <repo|org>
visibility: <public|private>
policy_profile: <strict|standard|light>
summary:
- repo_score: <0-100>
- risk: <low|medium|high>
- top_gaps: <count>
findings:
- id: G1
severity: <low|medium|high>
area: <metadata|community-health|templates|ownership|discoverability|governance>
observation: <what was observed>
expected: <what should be true>
gap: <delta>
evidence: <artifact/path/api result>
actions:
1) priority: <P1|P2|P3>
owner: <role or person>
change: <specific, minimal update>
rationale: <why this is sufficient>
verification: <objective pass condition>
decision:
- <apply|defer|NO_CHANGE>
missing_evidence:
- <none or required artifacts>
Evidence requirements
Use only relevant evidence for selected mode:
- repository metadata (
description, homepage, topics, visibility) - repository tree (
README,LICENSE, community files) .githubtemplates/config/code owners- applicable policy notes for exceptions
Stop conditions
Return NO_CHANGE when:
- selected mode lacks enough evidence,
- recommendation is not objectively verifiable,
- proposed change weakens safety or governance controls,
- same fix was recently applied and verified.
Quality bar
A valid output must be:
- specific (repo/path-level evidence),
- minimal (smallest effective change set),
- testable (clear pass/fail checks),
- traceable (finding → action → verification),
- repeatable (supports weekly re-check cadence).
What ships with it: 1 file
639 B alongside SKILL.md
- INSTALL-GLOBAL.md639 B
Gives 0 of the 12 instructions most audit compliance skills give in 927 tokens
Counted across 937 of the 1,487 authors here whose files we hold, read 2026-08-07
- Fetch latest guidelines before each reviewin 43 of 937, across 3 files
- Group findings by severityin 43 of 937
- Check files against all fetched rulesin 42 of 937, across 2 files
- Output findings in terse file:line formatin 41 of 937, across 3 files
- Ask user which files to review if none specifiedin 41 of 937, across 3 files
- Read specified files or prompt user for filesin 39 of 937, across 1 file
- Generate the audit reportin 33 of 937, across 30 files
- Assign a severity to every findingin 25 of 937
- Run automated accessibility scansin 23 of 937, across 13 files
- Output a markdown audit reportin 22 of 937
- Map findings to WCAG criteriain 20 of 937, across 10 files
- Confirm audit scopein 19 of 937, across 9 files
Said here and by no other author read
- limit one full repository pass per invocation
- limit ten recommendations per invocation
- produce explicit auditable proposals
- return NO_CHANGE if required artifacts are missing
- do not claim completion without verification evidence
- rank detected profile gaps by severity
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.