Skip to content
Back to skills

Qa

ASecurity

E2E and integration QA — derive the user flows a change affects and verify them (Playwright, HTTP, CLI) in a fix-verify loop. Use after implementation; unit tests → /tdd, /verify.

  • 2 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 2, 2026
ai-agentstypescriptgobashnodecode-reviewgitapisecurity

Works with

  • cli
  • api

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned September 2, 2026

npx -y skills add imoonkey/yaco --skill qa --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Qa?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Qa
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/imoonkey-yaco/badge)](https://www.skillsdirectory.com/skills/imoonkey-yaco)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: qa
description: E2E and integration QA — derive the user flows a change affects and verify them (Playwright, HTTP, CLI) in a fix-verify loop. Use after implementation; unit tests → /tdd, /verify.
---

# QA

Verify a change works from the user's perspective — E2E, integration, browser. Not unit tests.

| Skill | Level | What |
|-------|-------|------|
| `/tdd` | Unit | Write unit tests, RED→GREEN→REFACTOR |
| `/verify` | Gate | Build + lint + unit tests + security |
| **`/qa`** | **E2E / Integration** | **Verify user flows affected by a change** |

Triggers: after implementation, before a PR, or after deploy (smoke-test critical flows).

## Stack Detection

| Marker file | Stack | Reference |
|-------------|-------|-----------|
| `package.json` + web UI present | Web/Playwright | `references/web-playwright.md` |
| `package.json` + API/server | TypeScript/Node | `references/typescript-node.md` |

Read the matching reference for stack-specific commands.

## Process

### 1. Derive affected user flows

```bash
git diff --stat main...HEAD
```

From the diff, list the user flows (concrete actions a user takes) that touch the changed code, one assertion each:
- "User signs in → sees dashboard → data loads"
- "User runs `tool deploy --env staging` → exits 0, prints the deploy URL"
- "API receives POST /items → validates → returns 201"

### 2. Verify each flow

Pick the tool by surface:
- **Web UI** → Playwright: navigate, interact, assert
- **API** → HTTP: hit real endpoints, check responses
- **CLI** → run commands, check stdout/stderr/exit codes
- **Manual** → when automation isn't practical, read code + trace logic

Mark each flow **PASS** or **FAIL** (describe what broke).

### 3. Fix-verify loop

For each failure:
1. Categorize: **real bug** / **flaky** / **environment issue**
2. Fix real bugs only — one atomic commit per fix (`fix: ...`)
3. Re-verify the affected flow
4. Loop. Exit when all flows pass OR 3 consecutive fix failures.

- Flaky → note it, don't fix here.
- Environment issue → report and stop.
- On the 3-failure stop, report the flows still unverified.

### 4. Regression check

Re-verify **all** flows (not just fixed ones) to confirm no new breakage.

## Artifact

Write the result to a **sha-stamped artifact** in the project's `/yaco-paths`-resolved
bundle home (the same folder as the code-review artifacts), not just stdout. Stamp the
header with `reviewed_sha = <sha>`, the commit sha of the tree you exercised, so a gate
can confirm the QA covers the current HEAD by reading. Name it so it's recognizably a qa
artifact. The body is the Output Format below.

## Output Format

```
QA: [PASS/FAIL]
reviewed_sha: [sha of the tree exercised]

Flows verified:
- [flow description]: PASS/FAIL
- ...

Fixes:    [N commits]
Flaky:    [list or none]
Blocked:  [environment issues or none]
```

Files in this skill

  • SKILL.md2.8 KB
  • references/typescript-node.md1.6 KB
  • references/web-playwright.md1.4 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…