Audience MapperSAFE

120 marketing skills as an AI marketing staff — plugin, portable skills, or an 8-bot team across 7 disciplines (narrative, SEO/GEO, social, email, paid, influencer, launch) on one contract, with 8 auditor gates: TALE · CORE-EEAT · CITE · ECHO · SEND · ROAS · STAR · RAMP.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: audience-mapper
slug: audience-mapper
displayName: "Audience Mapper · 目标受众画像"
summary: "目标受众画像/人群分析 · 细分社群/亚文化调研"
description: 'Use when the user asks to "analyze my target audience", "build an audience profile for influencer targeting", "research a niche community", or "deep-dive a subculture before partnering with creators"; in audience mode produces demographic/psychographic profiles, a platform-priority matrix, named personas, and an influencer-selection criteria set, and in niche mode produces a community map, culture decode (language/norms/taboos), key-voice tiers, a Brand Fit Score, and a phased entry strategy. Not for finding specific creators to contract — use influencer-discovery; not for scoring a shortlist on Suitability — use fit-scorer. 目标受众画像/人群分析 · 细分社群/亚文化调研'
version: "20.1.0"
license: Apache-2.0
compatibility: "Claude Code and compatible agent-skill hosts"
homepage: "https://github.com/aaron-he-zhu/aaron-marketing-skills"
when_to_use: "Run at the start of an influencer program, or when entering a new market/segment, before any creator selection — this is the who + what-community step. Use audience mode to understand who the customer is, where they spend time online, which creators they trust, and what selection criteria follow; use niche mode to decode a specific subculture's language, norms, taboos, key voices, and brand fit before outreach so the brand avoids cultural missteps. From a brand or product name alone, produce hypotheses and a research plan; factual mapping requires supplied or fetched customer/community evidence. Also use to diagnose why a prior campaign underperformed or to build personas for a creative brief."
argument-hint: "<brand/product or niche> [mode: audience|niche] [category] [geo/platforms]"
metadata: {"author": "aaron-he-zhu", "version": "20.1.0", "discipline": "influencer", "phase": "scout", "geo-relevance": "low", "hermes": {"tags": ["marketing", "influencer", "scout"], "category": "influencer"}, "openclaw": {"emoji": "📣", "homepage": "https://github.com/aaron-he-zhu/aaron-marketing-skills"}}
---

# Audience Mapper

Maps **who** the brand is trying to reach and **what community** they belong to — the two halves of understanding an audience before any creator is selected. It runs in two modes against one shared inputs set:

- **`audience` mode** — the wide-angle read: demographic + psychographic profiles, a behavioral/media-diet map, a platform-priority matrix, content preferences, an influencer-affinity table, one or more named personas, and a must-have / nice-to-have / red-flag **influencer-selection criteria** set ready to hand to discovery.
- **`niche` mode** — the deep-dive: a community map (size, sub-niches, psychographics), a culture decode (language, norms, taboos), key-voice tiers, a content ecosystem, a **Brand Fit Score (X/25)** with a Strong/Moderate/Weak/Poor verdict, and a phased entry strategy with explicit red lines.

Both feed [STAR](../../../references/star-benchmark.md) creator/content scoring downstream, but this skill computes **neither** the Suitability/Trust/Appeal/Return dimension scores nor the SQS — it produces the audience and community facts that `fit-scorer` and `creator-content-auditor` later score against. Scope guard below.

## Quick Start

```
Analyze the target audience for [brand/product/category]          # audience mode
Build an audience profile for influencer targeting from this data: [data]
Research the [niche] community and identify opportunities for [brand]   # niche mode
Deep-dive [subculture] — key voices, what content works, brand fit, cultural risks
```

If the mode is not named, infer it: a broad brand/product/category request → **audience**; a named community, subculture, or hashtag (e.g. "#BookTok", "van-life") → **niche**. State which mode you picked before running.

## Skill Contract

**Expected output**: with sufficient dated evidence, in **audience** mode an audience analysis (demographics + psychographics, behavioral map, platform-priority matrix, content preferences, influencer-affinity table, ≥1 persona, and selection criteria); in **niche** mode a niche dossier (community map, culture decode, key-voice candidate tiers, Brand Fit Score X/25 + verdict, phased entry strategy, red lines). Without sufficient evidence, return hypotheses only as explicitly hypothetical, an exact research/collection plan, and `NEEDS_INPUT`; do not score, tier, promote, or hand off hypotheses as facts.

- **Reads**: the mode (audience / niche, inferred if unstated); brand or product name, category, geographic focus, price point, campaign objective; for niche mode the niche/community name, parent category, research goal (awareness/partnership/entry), and target platforms; any supplied first-party data (surveys, social insights, sales records, CRM). Prior `trend-spotter` or the sibling-mode's own output if present in `memory/influencer/`.
- **Writes**: return the mode-appropriate deliverable and reusable handoff inline by default; save to `memory/influencer/audience-mapper/YYYY-MM-DD-<topic>.md` only with exact authorization for that WARM path.
- **Promotes**: only with separate exact authorization, promote evidence-backed durable facts — in audience mode: target age range, priority platforms, ideal-influencer profile, persona ref(s); in niche mode: niche name, brand-fit verdict, top 3 `creator_ref` key-voice candidates, hard red lines/taboos — to `memory/hot-cache.md`. Hypotheses and unsourced current claims are never promoted.
- **Done when**:
  1. The chosen mode is stated, and every factual field carries source ref, observed/retrieved date, measurement window, and evidence label; confidence never converts an unsupported guess into fact.
  2. With sufficient evidence, complete the selected branch. Without it, return `NEEDS_INPUT` plus the exact queries/data fields; personas may be hypothetical but are not scored, tiered, promoted, or handed off as evidence.
  3. The deliverable and handoff are returned 

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.

[![ModelsAtlas audit: Audience Mapper](https://modelsatlas.com/skills/aaron-he-zhu/audience-mapper/badge.svg)](https://modelsatlas.com/skills/aaron-he-zhu/audience-mapper/)

Not your listing, or something out of date? Claim or update your listing.