# Playtest Evidence

> Use when planning or reviewing game playtests that need structured scenarios, participant context, observations, reproduction evidence, severity, and honest separation of observed findings from conclusions.

- Skill: `hoatv2211/playtest-evidence` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add hoatv2211/playtest-evidence`
- Raw SKILL.md: https://api.skillmd.com/api/skills/hoatv2211/playtest-evidence/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: hoatv2211 (https://skillmd.com/u/hoatv2211)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/hoatv2211/playtest-evidence

---

# Playtest Evidence

## Overview
Convert playtest notes into traceable evidence without turning opinions, memories, or small samples into invented metrics.

## When to use
Use for usability sessions, mechanic validation, onboarding tests, combat feel reviews, retention hypotheses, and regression playtests.

## When NOT to use
Do not use as a substitute for automated correctness tests, production analytics, or statistically representative research.

## Required inputs and context discovery
Collect build identity, feature hypothesis, participant context, scenario, environment, session method, consent constraints, observation timestamps, issue reproduction, and protected personal data rules.

## Safety and risk level
Read-only evidence synthesis. Remove personal data, do not record participants without consent, and label missing builds, recordings, or reproduction artifacts as `BLOCKED` or `Unverified`.

## Workflow
1. Bind the session to an exact build, feature hypothesis, and scenario.
   Completion criterion: the tested state and intended question are explicit.
2. Separate direct observation, participant quote, facilitator interpretation, and proposed fix.
   Completion criterion: every finding has one evidence class and source reference.
3. Normalize issues by reproduction, severity, frequency within the observed sample, and affected experience.
   Completion criterion: no issue uses unsupported population-level claims.
4. Compare results with the playtest criteria from the feature specification.
   Completion criterion: each criterion is Verified, Unverified, FAIL, or BLOCKED with a reason.
5. Produce follow-up experiments and regression checks.
   Completion criterion: each next step has an owner, artifact, and stopping condition.

## Evidence and output contract
Produce `playtest-evidence.json` with build identity, scenarios, participant context, observations, issues, criteria verdicts, limitations, and follow-up experiments.

## Handoff contract
Record session artifacts, privacy redactions, reproduced issues, unresolved disagreements, test gaps, next owners, and the build required for follow-up.

## Pitfalls and anti-rationalization
- A participant opinion is not an observed behavior metric.
- A small sample cannot justify a population percentage.
- Facilitator interpretation must not be rewritten as a participant quote.
- Missing recordings or builds remain explicit limitations.

## Verification checklist
- [ ] Build and scenario identities are recorded.
- [ ] Observations and interpretations are separate.
- [ ] Severity and frequency refer only to observed sessions.
- [ ] Privacy-sensitive data is removed.
- [ ] Conclusions cite evidence and limitations.

## References and scripts
Use the feature specification and project playtest artifacts. No live session runner is bundled; unavailable recordings or builds remain `BLOCKED`.