What does this skill do, and when should you use it?
Creator Registry is the repository’s protocol skill for maintaining creator-roster facts. It organizes pseudonymous creator aggregate IDs, verified handle links, rates, usage rights, exclusivity windows, compliance events, and performance baselines. It records provenance in an append-only creators event stream rather than scoring partner fit, reviewing content, or selecting creators. It fits teams that need controlled updates and an auditable source of truth for creator operations.
Reads creators.ndjson, its live projection, approved source records, and optional human views; queries the current state for a creator aggregate ID; accepts pending creator proposals; performs authorized owner updates and state transitions; regenerates a Markdown human view from accepted projection state; and runs creator-registry verification, reporting event IDs, offsets, revisions, conflicts, and expiring rights or exclusivity. Runtime operations use scripts/registry-events.py and Python 3.
- A creator-operations manager needs the latest rate, usage-rights, and exclusivity facts for a creator.
- A team needs to accept or reject pending creator proposals while preserving the decision history.
- An operator needs to deduplicate handles using verified cross-platform links or confirmed contact evidence.
- A campaign team needs to record closed-cycle rates and performance baselines with source and date.
- An organization needs an auditable record of creator-fact changes and their authorization.
How do you install this skill?
- Static files cannot confirm that registry-events.py, referenced protocols, capability signing, or projection verification actually run as described.
- Writes create persistent creator commercial and compliance records; reconfirm authorization, lawful basis, aggregate ID, and data minimization before the first persistent event.
- Erasure removes projected payload but leaves an audit tombstone and does not remove backups, Git history, or exported copies.
- Complete Chinese-language UX and mainland-China network reachability are not evidenced by this skill.
- Shell / CLI
- Local filesystem
Python 3Git
Clone the repository with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills, then expose the repository as a skill directory to an Agent Skills-compatible host. The skill is located at protocol/creator-registry/SKILL.md. The README documents installation for Claude Code and generic compatible hosts; it does not provide a creator-registry-specific single-skill command.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/protocol/creator-registry" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- What rate, rights, and exclusivity facts are current for creator-7f42?
- Accept or reject the pending creator proposals for creator-7f42.
Pass a creator aggregate ID or handle, for example: "What rate, rights, and exclusivity facts are current for creator-7f42?" or "Accept or reject the pending creator proposals for creator-7f42." Before any persistent write, confirm authorization and a lawful basis for natural-person data. A standalone folder may prepare proposals, but cannot append, project, or claim canonical creator truth without the verified root runtime, schema, and system catalog.
What are this skill's strengths and limitations?
- Append-only events preserve history, provenance, dates, authorization, IDs, and offsets.
- Pending proposals are kept separate from canonical accepted facts.
- Handle deduplication requires verification or user confirmation.
- Verification covers projections, conflicts, and expiring rights or exclusivity.
- Depends on the repository’s Python runtime, event schema, system catalog, and scripts.
- Accept, reject, update, and transition operations require the corresponding host capability.
- It does not score creator fit, audit submitted content, or choose partners.
- The supplied material does not show an independent test result or configured external connector for this skill.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Creator Registry this page | 61 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| Launch Registry | 59 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| Channel Registry | 58 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| Consent Registry | 62 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| Dependency Refresh Playbook | 59 · Recommended | ★ 4.5k | 23d ago | MIT |
How did FollowSkills review this skill?
The skill specifies data minimization, pseudonymous aggregate IDs, explicit user authorization and lawful basis, proposal-only writes, host capabilities, revision checks, append-only history, and erasure tombstones; 7 points are deducted because capability deployment depends on an external host security boundary, runtime behavior was not verified, rollback is not demonstrated, and erasure does not cover backups or exports.
The instructions provide dependency checks, Unknown-state handling, stale-revision reconciliation, conflict preservation, pending behavior when authority is unavailable, and a verification step, making the visible happy path fairly coherent; 1 point is deducted because the referenced protocols, scripts, catalog, and runtime are not fully supplied and were not executed, leaving abnormal-path behavior incompletely evidenced.
Triggers, aggregate-ID/handle inputs, query/write boundaries, and separation from fit-scoring and content-auditing are clear, with compatibility stated for Claude Code and compatible hosts; 4 points are deducted because Chinese-language interaction is only lightly evidenced and reachability in mainland-China networks is not established.
The skill has structured frontmatter, versioning, Apache-2.0 licensing, quick-start examples, contract sections, references, a template, limitations, and next-skill routing; 4 points are deducted because the skill lacks an explicit changelog, named maintenance responsibility, and update path, while publisher identity is unverified by the curated registry.
The skill defines querying creator facts, proposal acceptance, deduplication, recording, projection generation, and completion verification, with expected event IDs, offsets, and revisions, so the core task appears directly usable; the static calibration cap limits this to 7 because no executed results or representative end-to-end outputs verify usability or incremental benefit over manual work.
The supplied material includes repository regression tests, audit-validator tests, a security policy, and deterministic event/projection design, providing limited auditable evidence; the static calibration cap limits this to 5 because the tests do not demonstrate coverage of this skill's key paths and there is no independent third-party reproduction evidence.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →