What does this skill do, and when should you use it?
Channel Registry is a protocol-layer skill in the repository for maintaining facts about brand-owned social channels, UGC permissions, advocate opt-ins, cadence commitments, and platform voice adaptation. It records state through an append-only channel event stream and regenerates channel, voice, permission, roster, and cadence views from accepted projections. It does not perform ECHO scoring or choose the channel portfolio. It fits teams that need explicit ownership, permission scope, expiry dates, provenance, and revision history.
Reads the channels event stream, current projection, approved narrative and rights evidence, platform-rule snapshots, and cadence commitments; creates or updates channel events through registry-events.py using host-capability owner-append and the current expected_revision; manages proposed, warming, active, paused, and retired lifecycle states; records UGC permission scope, channels, duration, compensation, and expiry; accepts or rejects activity and incident proposals submitted by other social skills; regenerates channel, voice, permission, advocate-roster, and cadence views; and runs verify channels. Completed work should include revisions, provenance, accepted or rejected events, conflicts, and a next-skill handoff.
- A social lead needs to register a new account with its control, governance, objective, narrative pointer, and cadence evidence.
- An operations team needs to move a LinkedIn account from warming to active while preserving graduation evidence and revision history.
- A brand receives user-generated content and needs to record organic-only permission with approved channels and an expiry date.
- A team needs to accept or reject pending social-activity or incident proposals from listening, inbox, or crisis producers.
- A governance owner needs to query current channel state, voice adaptation, permission scope, and advocate opt-in evidence.
How do you install this skill?
- Do not perform writes or claim canonical channel state until AARON_SKILLS_ROOT, registry-events.py, the event schema, and the system catalog are verified.
- Owner operations must be confirmed by an authorized host-capability principal; request actor fields do not substitute for authority.
- A public post, tag, or hashtag is not UGC permission; verify scope, channels, duration, compensation, and expiry evidence.
- Resolve the governance mismatch between skill version 18.0.0 and the Security Policy's supported 17.x line, and add runtime, skill-specific tests, and clearer rollback documentation.
- Shell / CLI
- Local filesystem
Gitregistry-events.py
Install the repository with: git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The skill is located at protocol/channel-registry/SKILL.md. In Claude Code, the collection can also be installed with /plugin marketplace add aaron-he-zhu/aaron-marketing-skills followed by /plugin install aaron-marketing@aaron. The README does not provide a clearly documented single-skill installation command for this skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/protocol/channel-registry" ~/.claude/skills/
rm -rf "$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:
- Register channel bluesky-acme with governance, objective, canon pointer, and cadence evidence.
- Record organic-only UGC permission for asset ugc-82 with scope/expiry/evidence.
After installing the skill in an Agent Skills-compatible host, issue a concrete request such as: “Register channel bluesky-acme with governance, objective, canon pointer, and cadence evidence.” or “Record organic-only UGC permission for asset ugc-82 with scope/expiry/evidence.” Before invoking the runtime, read registry-event-protocol.md and runtime-invocation.md and verify the registry script, event schema, and system catalog. A standalone folder installation may prepare proposals, but cannot append or project events or claim canonical channel state without a verified root runtime, schema, and catalog.
What are this skill's strengths and limitations?
- Uses an append-only event stream for traceable channel state, decisions, revisions, and provenance.
- Separates channel lifecycle, platform voice adaptation, UGC permissions, and advocate opt-ins.
- Captures permission scope, channels, duration, compensation, and expiry instead of treating organic permission as paid usage rights.
- Provides structured handoffs with current state, conflicts, accepted or rejected events, and a next skill.
- Does not score ECHO or select a channel portfolio; those tasks require other skills.
- Depends on a verified root runtime, event schema, system catalog, and registry-events.py.
- Requires local filesystem and shell access; standalone installation cannot append, project, or claim canonical channel state.
- The supplied source does not document a specific test suite, supported platform list, or external platform connectors 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 |
|---|---|---|---|---|
| Channel Registry this page | 58 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Launch Registry | 59 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Creator Registry | 61 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Consent Registry | 62 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Discord Channel Access ✓ Anthropic · Official | 53 · Use with care | ★ 38k | 12d ago | Apache-2.0 |
How did FollowSkills review this skill?
The skill requires explicit authorization, host capability, append-only events, scoped and expiring permissions, and minimization of personal data; it also forbids treating public content as UGC authorization. These support strong least-privilege and governance practices. Eight points are deducted because the required runtime and protocol files are not supplied, rollback is only partial, runtime memory is not encrypted, and publisher identity is unverified.
The skill specifies projection reads, expected_revision compare-and-set updates, a lifecycle state machine, pending-proposal fallback, and verification. The review is static, with no skill-specific tests covering key paths and no concrete failure-message examples; it also depends on omitted runtime and reference files. The score is therefore limited to 8.
Triggers, units, non-fit boundaries, evidence inputs, completion criteria, and handoff fields are reasonably clear, with Chinese channel-registry terminology and a standalone-install boundary. Points are deducted for assumptions about Claude/git and a verified root runtime, abstract argument syntax, and no evidence of platform or mainland-China reachability. Score: 11.
The file has stable naming, slug, version, Apache-2.0 licensing, progressive links, Quick Start, data sources, save rules, and next-skill guidance. Points are deducted because no skill-specific changelog, maintenance owner, or update path is shown; the Security Policy says 17.x is supported while the skill and README identify 18.0.0, creating governance inconsistency.
The objective is clear, and the event stream plus regenerated views should produce channel, permission, cadence, and handoff records usable by downstream social workflows. No real execution result or representative output is supplied, and core runtime files are omitted, so the static ceiling of 7 applies; execution correctness and completeness remain unverified.
The repository supplies general architecture, audit-artifact, and PII tests plus textual integrity claims, providing limited auditability. It does not supply channel-registry-specific tests, execution records, third-party review, or the referenced protocol contents. Coverage and independent reproducibility are therefore insufficient for more than 5.
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 →