Audience Mapper
Map the people and communities to understand before creator selection.
The skill distinguishes user-supplied data from connector data, treats exports and fetched content as untrusted, and requires confirmation before writing durable memory. Its write scope, connector role, and downstream handoffs are disclosed. Points are deducted because data minimization, sensitive marketing-data handling, actual connector permissions, and rollback depend largely on shared documents not supplied here, while a brand or community name alone can lead to substantial inference.
The two modes, steps, completion criteria, templates, and scope guard are broadly self-consistent, and inferred attributes must carry confidence labels. Points are deducted because static review cannot reproduce key paths; handling of missing data, ambiguous modes, conflicting sources, and unverifiable community facts lacks concrete diagnostic failure behavior, and referenced shared protocol files were not provided. The static cap keeps this at 9/20.
Audience and niche modes have clear trigger language, inputs, outputs, and non-fit boundaries, making the skill suitable for early influencer-program research. Points are deducted because the skill permits broad demographic and cultural inference from minimal input, provides limited evidence for boundary validation, and its platform templates focus on Instagram, TikTok, YouTube, and X without clearly specifying mainland-China alternatives; geo-relevance is also marked low.
Frontmatter supplies the name, version, license, compatibility, triggers, and argument hint. The documentation uses progressive sections for Quick Start, Contract, Data Sources, Instructions, Templates, and Next Skill, with limitations, save flow, examples, and repository security material. Points are deducted because the skill does not clearly identify a maintainer, changelog, or update path; important shared references were not included in the supplied evidence, and examples are mostly templated rather than real outputs.
The contract defines directly handoff-ready audience profiles, platform matrices, personas, influencer criteria, niche dossiers, Brand Fit Scores, and phased strategies. The scope guard and downstream handoffs are clear. Points are deducted because results depend heavily on user data or external research, and key voice, size, engagement, and platform-performance fields can remain unsupported placeholders or inferences. Static calibration limits this to 7/15 and cannot establish execution quality.
The supplied files provide explicit inputs, outputs, confidence labeling, safety boundaries, Apache-2.0 licensing, and repository-level CI/test evidence, reaching the static-review ceiling. Points are deducted because there are no skill-specific execution records, key-path tests, cited examples, or independent corroboration; the repository tests do not demonstrate audience-mapper behavior. Static calibration limits this to 5/10.
- Do not treat demographic, psychographic, platform, key-voice, or cultural-taboo claims generated from only a brand or community name as facts; require sources, confidence labels, and human validation before consequential decisions.
- For mainland-China audiences, add reachable local platforms, language handling, and local data sources; the supplied templates provide weak evidence of that fit.
- Confirm connector permissions, data-egress scope, sensitive-data handling, and memory recovery behavior in the actual host; this review was static only.
What does this skill do, and when should you use it?
Audience Mapper is the influencer-scout audience research skill in Aaron Marketing Skills. It offers an audience mode for broad customer and creator-targeting analysis, and a niche mode for decoding a specific community or subculture. Its outputs include profiles, platform priorities, content preferences, key-voice tiers, brand fit, and cultural risks. It is a fit for teams that need audience clarity before outreach, but it does not find, contract, or score specific creators.
Reads the supplied brand or product, category, geography, price point, objective, platforms, and first-party customer or community data, then selects or infers the audience or niche mode. Audience mode produces confidence-labeled demographic and psychographic profiles, behavioral patterns, a platform-priority matrix, content preferences, influencer affinity, named personas, and creator-selection criteria. Niche mode produces a community map, culture and taboo analysis, tiered key voices, a content ecosystem, a Brand Fit Score out of 25, and a phased entry strategy. After confirmation, results may be saved under memory/influencer/audience-mapper/ and durable facts promoted to memory/hot-cache.md.
- An influencer team is starting a program and needs to define its target customers, platforms, and creator trust factors.
- A marketing team is entering a new niche or subculture and needs its language, norms, taboos, and key voices decoded.
- A creative team needs named audience personas for an influencer brief.
- A team is diagnosing why a previous creator campaign underperformed.
What are this skill's strengths and limitations?
- Provides distinct broad-audience and deep-niche research modes.
- Produces concrete outputs such as platform matrices, personas, creator criteria, and a Brand Fit Score.
- Runs from user-supplied data without requiring live connectors.
- Clearly separates audience mapping from creator discovery and creator-fit scoring.
- Output quality depends on the evidence supplied; inferred attributes must carry confidence levels.
- It does not discover or contract specific creators or score a shortlist for Suitability or SQS.
- Persistent memory writes require user confirmation.
- The supplied SKILL.md provides no independent test results or platform-validation report for this individual skill.
How do you install this skill?
On an Agent Skills-compatible host, run: npx skills add aaron-he-zhu/aaron-marketing-skills -s audience-mapper. In Claude Code, you can instead run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, followed by /plugin install aaron-marketing@aaron. The repository is licensed under Apache-2.0.
How do you use this skill?
Use a prompt such as: Analyze the target audience for [brand/product/category], or Research the [niche] community and identify opportunities for [brand]. You may also provide: <brand/product or niche> [mode: audience|niche] [category] [geo/platforms]. If no mode is stated, broad brand or product requests infer audience mode, while named communities, subcultures, or hashtags infer niche mode.
How does this skill compare with similar options?
Unlike influencer-discovery, this skill defines the audience, community, and selection criteria before sourcing creators. Unlike fit-scorer, it does not score a creator shortlist on Suitability.