Influencer Fit Scorer
Evaluate shortlisted creators with an evidence-based suitability read and a separate campaign-priority ranking.
The skill requires explicit authorization before writing reports and keeps commercial fit separate from Suitability; no malware, covert exfiltration, or destructive default is shown, and shared security material describes non-persistent connector keys. Deducted 10 points because the skill itself does not fully explain sensitive audience/creator data handling, end-to-end data flows, revocation or rollback, and optional external fetching/API use still creates permission and privacy exposure. Publisher identity is unverified.
The workflow, inputs, outputs, Unknown handling, and separation of scoring surfaces are largely internally consistent, with repository-level CI and artifact-validator evidence. Deducted 12 points because static review cannot reproduce key paths; the core STAR rubric, shared contract, and connector implementations are not supplied; and the YouTube connector sits somewhat uneasily beside the claim that live integrations are unnecessary. Failure feedback for abnormal inputs is also limited.
Triggers, required inputs, output boundaries, and non-fit cases for discovery and outreach are clearly stated, with Chinese terminology present. Deducted 5 points because platform, metric, and market boundaries are incomplete; geo-relevance is marked low; and mainland-China reachability, YouTube/API availability, and cross-platform operation are not specifically established.
The skill has frontmatter, versioning, license metadata, Quick Start prompts, contract, data sources, numbered instructions, templates, examples, and handoff guidance. Deducted 6 points because maintenance ownership and update path are unclear, key referenced materials are absent from the supplied evidence, examples remain heavily templated, and SECURITY.md supports 17.x while the skill and repository identify 18.0.0.
The evidence indicates a useful intended workflow: record S1-S10 per creator and produce a separately labeled commercial matrix for prioritization. Deducted 9 points because no skill-specific tests or verified representative outputs are supplied; the core Suitability result depends on the unavailable STAR rubric and external evidence; Unknown can prevent a complete read; and the templates require substantial manual completion.
The skill requires dated sources, evidence types, confidence, and gap reasons, while repository materials include architecture and audit-artifact tests. Deducted 6 points because there are no skill-specific key-path tests, third-party execution records, or independent corroboration, and the supplied material omits the STAR benchmark and several referenced documents needed to verify the central scoring logic.
- This is a static source review only; the skill, connectors, and tests were not executed. Repository-level CI evidence must not be treated as validation of this skill's key paths.
- Core Suitability conclusions depend on the unsupplied STAR benchmark and external creator evidence. Missing data must remain Unknown and should not be converted into Pass or a definitive ranking.
- Before using YouTube or other social-data sources, confirm API-key handling, platform terms, privacy authorization, mainland-China reachability, and explicit user approval for data flows.
- Note the version-governance mismatch between skill/repository version 18.0.0 and SECURITY.md support for 17.x.
What does this skill do, and when should you use it?
Fit Scorer is for brands and marketing teams that already have a creator shortlist. It evaluates the STAR Suitability (S) dimension and records each S1–S10 item as Pass, Partial, Fail, Unknown, or N/A. Campaign-specific factors such as budget, rates, availability, and partnership potential remain in a separately labeled commercial-fit matrix rather than entering the Suitability score. A formal Suitability read requires complete applicable coverage with dated evidence; Unknown prevents the read from being issued.
It reads brand or campaign context, target-audience definitions, campaign goals, shortlisted creator handles, and optionally audience profiles, competitor benchmarks, and roster records. It locks the typed assessment context, freezes evidence with source, date, type, and confidence, and evaluates S1–S10 under the STAR benchmark, covering audience composition, follower-growth integrity, reach reliability, engagement health, credibility, and portable brand or category fit. It can build a separate commercial-fit comparison, report critical-control evidence, confidence, and outreach recommendations, and save a report only after explicit authorization. YouTube candidates can use the repository's Python connector for measured inputs.
- A brand has TikTok, Instagram, or YouTube candidates and needs a consistent shortlist-ranking process.
- A marketing lead must compare creators and justify a selection to internal stakeholders.
- A campaign team wants weighted evaluation standards for awareness, engagement, or conversion goals.
- A brand needs to separate a creator's portable suitability from campaign-specific cost, availability, and partnership potential.
- A team is building consistent long-term creator-partner tiers from documented evaluations.
What are this skill's strengths and limitations?
- Keeps portable Suitability analysis separate from campaign-specific commercial fit.
- Requires explicit S1–S10 states with evidence source, date, type, and confidence.
- Treats missing evidence as Unknown instead of silently converting it into Partial or Fail.
- Works with user-supplied data without requiring external integrations.
- Defines permissioned report persistence and a clear handoff to the STAR auditor.
- Does not discover new influencers or conduct outreach.
- Users must supply the shortlist, target definition, and sufficient evidence; Unknown blocks a formal Suitability read.
- The commercial-fit score is not a STAR Suitability score and cannot override a Suitability veto or evidence gap.
- The YouTube measurement connector references Python 3 and requires YOUTUBE_API_KEY when used; other platform data depends on user inputs or optional connectors.
- The source does not provide an independent test suite or platform-coverage validation for this individual skill.
How do you install this skill?
Install the repository as a collection. In Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then /plugin install aaron-marketing@aaron. On an Agent Skills-compatible host, run npx skills add aaron-he-zhu/aaron-marketing-skills, or clone it with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The skill is located at influencer/scout/fit-scorer/SKILL.md. The source does not document a dedicated single-skill installation command for Fit Scorer.
How do you use this skill?
Provide the brand or campaign context, target audience, goal, shortlisted handles, and available evidence. For one creator, use: Score @[handle] for [brand/campaign] and tell me if they're a good fit. For a shortlist, use: Compare and rank these influencers for [campaign]: @influencer1, @influencer2, @influencer3. If typed-context fields are missing, the skill should return NEEDS_INPUT and name them. Use influencer-discovery for sourcing creators and outreach-manager for contacting them.
How does this skill compare with similar options?
Unlike influencer-discovery, Fit Scorer does not source or expand a creator list; it evaluates an existing shortlist. Unlike outreach-manager, it does not contact creators, manage follow-ups, or negotiate rates.