agentleFS
mtarcuremtarcure/claude-vibe-squad/.agents/skills/rule6-rights-gate/SKILL.md

rule6-rights-gate

Skill220 starsChanged 20 days agoMITView on GitHub

Hard Rule 6 pre-publication rights gate — check licensing/provenance AND likeness/voice consent over a generated or third-party media asset before publish/use, and emit the machine-readable gate record. Use at the S4 Verify step of project/image, project/video, project/audio-assets (or any asset headed for publication).

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    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
  1. Distinctive:

    No infringement asserted from similarity alone.

    line 71

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Rule-6 Rights Gate

Fuses rights-and-provenance-gate + consent-and-likeness-check into one invokable pre-publication gate over a media asset. It binds the decision to an immutable asset hash, checks licensing/provenance + real-person likeness/voice consent, and emits a machine gate record. It uses non-legal language and never adjudicates fair use / de-minimis; a material unresolved check is a HOLD, not a PASS.

Procedure

  1. Bind the decision to an immutable subject_id / subject_hash / subject_version and a versioned checklist. A modified asset (new hash) requires a NEW gate record — a prior PASS does not carry over.
  2. Provenance + licensing, each with evidence: license/provenance, usage-terms fit (channel / territory / duration), trademark, watermark / C2PA markers, and music match. Record "identified match/similarity" in non-legal language; do NOT decide de-minimis or fair use, and never assert infringement from similarity alone.
  3. Likeness + consent: detect real-person likeness — voice, face, or persona. Require documented consent scope on file, OR confirm the asset is synthetic-generic (no identifiable person). Confirm the intended use (channel / territory / duration) falls within the consent granted.
  4. HOLD on any material unresolved check — a lookup/verification tool not wired, an unconsented real-person likeness, an unresolved license — is HOLD, never PASS on uncertainty. Do not self-clear a likeness.
  5. Route PII / biometric processing to privacy-steward (likeness work may need both; neither is legal counsel). Genuinely legal-uncertain items surface to a human / counsel, not adjudicated here.
  6. Emit the gate record (see below).

Machine gate record (emit this)

rule6_rights_gate:                   # this block key names the gate type (was gate_type=rights)
  result: PASS | HOLD | FAIL
  gate_version: <schema/checklist version>   # folded from asset-provenance-and-rights-auditor
  subject_id: <stable asset id>      # folded in; identifies the asset across versions
  subject_hash: <hash>               # decision is bound to this exact asset version
  subject_version: <version>         # folded in; the asset revision this decision covers
  assurance: <full | restricted>     # restricted when reverse-image / registry / C2PA lookups are unavailable
  checklist:                         # each item: evidenced | held
    license_provenance: <...>
    usage_terms_fit: <...>
    trademark: <...>
    watermark_c2pa: <...>
    music_match: <...>
    likeness_consent: <...>
  evidence_refs: [<refs>]            # folded in; evidence backing the checklist items
  unresolved: [<items>]              # any material item here forces HOLD
  privacy_handoff: <yes|no>          # yes if PII/biometric processing is involved
  specialist: <role>                 # folded in; who ran the gate
  reviewer: <role>                   # who reviewed it
  timestamp: <iso>                   # completion time (was completed_at)
  override: <actor + reason, if a human override was recorded>   # holds override_actor + override_reason

Single source. This is the one schema for the Rule-6 gate record; departments/content/specialists/asset-provenance-and-rights-auditor.md references it rather than restating it (one fact, one home). gate_type is dropped as redundant — the rule6_rights_gate: block key already names the gate.

When to invoke

  • The S4 Verify step of project/image, project/video, project/audio-assets — any asset (generated or third-party) headed for publication/use.

When NOT to invoke

  • Purely internal, non-published scratch assets with no real-person likeness and no third-party material — but when in doubt, run it.
  • As legal advice — it produces a risk record for a human/counsel, not a legal determination.

Acceptance

  • Decision is hash-bound; every checklist item is evidenced or HELD. No PASS on a material unresolved check.
  • No infringement asserted from similarity alone; legal uncertainty is surfaced, not adjudicated.
  • Any real-person likeness has consent on file or is HELD; consent scope matches intended use; privacy/biometric handoff made where processing is involved.

Notes — supersession + cross-lane home

  • Sources fully subsumed: the retired rights-and-provenance-gate and consent-and-likeness-check skills. Every component step and acceptance condition is represented by the combined procedure and record above; the standalone skills are retired (shared/skills/_retired/), not parallel S4 triggers.
  • Canonical home: .claude/skills/rule6-rights-gate/SKILL.md. Because its audience is specialist, model-lanes/SKILL-HOMES.md requires .agents/skills/rule6-rights-gate/SKILL.md to be a byte-identical regular file mirror.
  • Distinct from privacy-steward (PII/data-flow) and asset-provenance-and-rights-auditor (the heightened-risk role that owns clearance) — this skill is the media S4 self-gate + record emitter.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.