break-test

Files (1)

• SKILL.md

SKILL.md (first lines)

# Break Test Skill

Input/context:

```text
$ARGUMENTS
```

## Mission

Find the ways this fails before a buyer does.

## Attack surfaces

Evaluate:

1. **Technical failure** — crashes, bad data, flaky dependencies, latency, missing env vars.
2. **AI failure** — hallucination, prompt injection, non-determinism, overclaiming, brittle parsing.
3. **Workflow failure** — buyer cannot install, run, trust, or interpret it.
4. **Commercial failure** — weak ROI, no urgency, unsupported pricing.
5. **Distribution failure** — no reachable channel, no credible wedge.
6. **Support failure** — too much human handholding, unclear errors, hidden manual steps.
7. **Trust failure** — security, privacy, compliance, auditability, reversibility.

## Output

Return:

- top 10 failure modes
- severity / likelihood matrix
- what evidence would reveal each failure
- fixes ranked by leverage
- what can be ignored for MVP
- final verdict: `SHIP TO TEST BUYER`, `FIX FIRST`, or `RETHINK`

## Durable output

If working in a repo, create or update:

```text
docs/profit/break-test-report.md
```

From az9713/idea-to-capital, MIT.

More from az9713/idea-to-capital