What does this skill do, and when should you use it?
Creator Content Auditor is the Influencer/Activate audit skill in the Aaron Marketing Skills repository, designed for pre-publication creator-content quality checks. It reads the submitted asset, campaign brief, approved claims and disclosures, platform requirements, and creator-fit evidence, then evaluates the STAR framework. It returns a profile-weighted SQS (Star Quality Score), explicit audit status and verdict, and constructive creator revision feedback. It does not draft briefs, adjudicate claims, or manage partnership terms.
Locks the submission, brief or canon version, market, platform, campaign goal, and assessment_time; reads the fit-scorer Suitability result, creator-registry dossier, claim substantiation, disclosure facts, and Return evidence; scores applicable STAR items S1–S10, T1–T10, A1–A10, and R1–R10, marking R1–R6 as na for forecast assessments; checks disclosure, factual-claim, brand-safety, follower-fraud, and fake-engagement vetoes; runs python3 "$AARON_SKILLS_ROOT/scripts/rubric-score.py" score <run.json> when the verified runtime is available; and produces an SQS audit report plus location- or timecode-specific creator revision feedback.
- A brand team reviews sponsored video and caption before amplification or a payment milestone.
- A campaign manager needs a go/no-go decision against an approved brief and claims record.
- A compliance or brand-safety lead checks commercial disclosure, product claims, and authenticity risks.
- A creator submits a revision and the team needs a traceable rerun against the same brief or canon version.
How do you install this skill?
- This is a low-confidence static review; the runtime, scripts, validator, and scorer were not executed.
- Before use, verify all parent references and runtime components, and reconcile the conflicting FIX versus Reject semantics for a single STAR-T1 veto.
- Chinese users may need a Chinese audit template; platform, FTC/ASA, and market requirements require human review for the target jurisdiction.
- Shell / CLI
- Local filesystem
Python 3Git
Install the repository collection first: in Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills and then /plugin install aaron-marketing@aaron; on other Agent Skills-compatible hosts, run npx skills add aaron-he-zhu/aaron-marketing-skills, or use git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The source also supports single-skill installation on generic hosts, but it does not provide a dedicated command for this skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/influencer/activate/creator-content-auditor" ~/.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?
Place influencer/activate/creator-content-auditor/SKILL.md in an Agent Skills-compatible host, or trigger it through the installed collection. Example: review a sponsored video and caption against campaign brief v4 for conversion, run the STAR gate, show claim and disclosure blockers, return the SQS, and write the creator revision note. Provide the frozen submission, brief or canon version, approved claims and disclosures, platform requirements, Suitability result, creator dossier, market, goal, and assessment_time.
What are this skill's strengths and limitations?
- Clear ownership: this skill is the sole calculator of the STAR SQS.
- Checks hard vetoes covering disclosure, factual claims, brand safety, follower fraud, and fake engagement.
- Treats missing qualified evidence as Unknown instead of converting it into N/A or failure.
- Requires feedback to map each change to a location or timecode, problem, correction, example, owner, and resubmission condition.
- Supports forecast and actual Return assessment times.
- It requires qualified inputs such as the brief, claim substantiation, platform requirements, and creator-fit evidence; missing evidence prevents scoring and a gate verdict.
- The runtime references Bash/Python and local files; complete scoring depends on the repository's scorer, validator, and typed catalogs.
- The source provides no platform-coverage list, test-suite results, or independent platform testing evidence.
- It does not adjudicate claims or rights and does not draft briefs, contracts, or creator content.
- Writing an audit artifact requires permission and the specified v3 validation and persistence flow.
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 Content Auditor this page | 56 · Use with care | ★ 3k | 12d ago | Apache-2.0 |
| Creator Contract Helper | 58 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
| Influencer Fit Scorer | 52 · Use with care | ★ 3k | 12d ago | Apache-2.0 |
| Influencer Discovery | 51 · Use with care | ★ 3k | 12d ago | Apache-2.0 |
| Social Quality Gate | 62 · Recommended | ★ 3k | 12d ago | Apache-2.0 |
Unlike brief-generator, it audits submitted creator content rather than producing an influencer brief. Unlike contract-helper, it does not handle partnership terms or usage rights. It also depends on fit-scorer for Suitability and can use roi-calculator evidence for an actual Return reread.
How did FollowSkills review this skill?
The evidence shows explicit treatment of submissions, metadata, QR codes, and embedded instructions as untrusted data, user permission before writes, version locking, read-only review behavior, and constrained persistence. However, the skill still depends on several runtimes, registries, and platform materials without fully specifying sensitive-data handling, dependency isolation, or recovery when unavailable; 5 points are deducted.
The document defines inputs, states, missing-evidence handling, a NOT_SCORED failure path, and feedback requirements, giving the workflow a plausible deterministic structure. However, parent references, scorer, validator, and runtime availability are not fully verifiable from the supplied material, and no dedicated tests cover this skill's key paths. The static ceiling applies; 12 points are deducted.
Triggers, audience scenarios, required inputs, SQS and creator-feedback outputs, non-fit cases, and four goal profiles are relatively clear. Chinese execution and output support is limited, platform/market handling depends on external notes, and geo-relevance is explicitly low; 5 points are deducted.
The skill has structured front matter, Quick Start, Skill Contract, data-source table, validation checkpoints, persistence rules, references, versioning, and an Apache-2.0 license. Maintenance ownership, update path, and changelog coverage are incomplete, while SECURITY.md identifies 17.x as the current line despite the skill being 18.0.0; 5 points are deducted.
The skill targets pre-publish auditing, disclosure and claim vetoes, SQS calculation, and concrete revision feedback, with templates, checklists, and a worked example. However, key runtime and reference availability cannot be confirmed statically, and the templates describe a single disclosure veto as Reject while the main rules/examples allow a single veto to produce FIX, creating possible output inconsistency; 10 points are deducted.
A versioned typed framework snapshot, source digest, controlled state vocabulary, and auditable fields provide limited traceability. There is no independent execution evidence or multi-source corroboration for this skill's key paths, and the supplied CI material does not demonstrate coverage of it; 7 points are deducted.
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 →