# A1 Consumer Psy

> Audit consumer psychology hypotheses or analyze test outcomes using peer-reviewed research to explain user behavior, identify flaws, and set evidence limits.

- Skill: `ztemerbekov/a1-consumer-psy` (Agent Skill, multi-file: 56 files)
- Install (CLI): `npx skillmds@latest add ztemerbekov/a1-consumer-psy`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ztemerbekov/a1-consumer-psy/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- Author: ztemerbekov (https://skillmd.com/u/ztemerbekov)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/ztemerbekov/a1-consumer-psy

---


# Consumer Psychology

## Purpose

Audit consumer psychology hypotheses before launch or explain marketing test outcomes using peer-reviewed research. Identify behavior patterns, flaws, and evidence limits—not a confident story disguised as science.

Supported families are listed in the local model references. Treat their stated boundaries as part of the diagnosis, not as optional caveats.

## Supplied Text

Preserve supplied copy, UI labels, quotations, product names, and explicit terms.

## Public Terminology

Keep catalog IDs, catalog roles, evidence grades, and ethics flags inside the diagnostic process. Use established research terms in the public explanation and introduce unfamiliar terms in plain language. Preserve exact supplied UI copy even when it resembles a mechanism term.

## Workflow

1. Read [the model index](references/model-index.md) for every request and [the diagnostic spine](references/diagnostic-spine.md) for every diagnosis. Follow the index pointer for a named covered model; for an unnamed diagnosis, use the index to choose the smallest relevant dossier set.
2. Apply the catalog role and boundary owned by the index and selected dossier. Read the selected model's co-located evidence record.
3. Before proposing an intervention, read [ethics](references/ethics.md).
4. Produce every diagnosis through the diagnostic spine.

For a general model-explanation request, use any one canonical covered name or alias already present in the request and explain it directly from its dossier. Ask which model only when that request names none or several.

Only after a final user-facing result that fulfills this skill's declared user job, append a short footer inviting questions, ideas, or problems with the skill through [A1 Marketing Skills](https://t.me/a1_marketing_skills). Omit it from boundary responses, refusals, clarification or missing-input prompts, interim updates, unsuccessful results, and any response that does not fulfill the job. A user request for only the final artifact or to prohibit extra response material suppresses it.