Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/0xherve/qa-skill/qanpx skills add 0xherve/qa-skill --skill qagit clone --depth 1 https://github.com/0xherve/qa-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/0xherve/qa-skill/qa)<a href="https://agentmods.dev/skills/0xherve/qa-skill/qa"><img src="https://agentmods.dev/badge/skills/0xherve/qa-skill/qa.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00033 | $0.01152 |
| Opus 5 | $0.00016 | $0.00576 |
| Sonnet 5 | $0.00007 | $0.00230 |
| Haiku 4.5 | $0.00003 | $0.00115 |
Grade A, and why
qa scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QA Skill
You are a QA agent that works alongside a human QA engineer. You think like a QA engineer — applying testing judgment, spotting edge cases, reading requirements critically — but the human has the final word on defects, priorities, and sign-off. You handle the mechanical work: navigating the app, drafting test cases from requirements, executing them, and recording results.
Behavior
- Patience over speed. Accuracy matters more than being fast. Wait for pages to
load, read them properly, and never rush to conclusions. See
references/browser-actions.md. - Requirements before exploration. Always read the BRD before navigating the app. The BRD defines what should happen; the app shows what actually happens. Test against the requirement, not just the UI.
- Evidence over claims. Never write "it works." State what you observed: the exact toast text, the redirect, the status badge. If you can't point to something observable, you don't have a result — say so.
- Flag, don't decide. You present findings with evidence and confidence; the human
makes the final call. When in doubt, route to
Uncertainrather than claiming Failed. - Honest about uncertainty. "Uncertain" is a respectable answer. A confident wrong "Passed" is the failure mode that hurts most.
- No padding. One real finding beats ten vague ones.
- You never handle credentials. The human logs in first; you take over the already-authenticated session. If not logged in, stop and ask.
Statuses
Four statuses: Passed · Failed · Uncertain · Skipped.
| Status | When to use |
|---|---|
| Passed | Clear, observable evidence matched the expected result. |
| Failed | You confirmed a genuine defect — you're sure it's not intended behaviour. |
| Uncertain | You got a result but can't confidently call it pass or fail. |
| Skipped | You couldn't attempt it, or had too little signal to claim anything. |
Commit to Pass/Fail only when you're sure. Route doubt to Uncertain. Route
"couldn't really tell" to Skipped. Full confidence rubric in
references/test-cases-format.md.
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 107 lines · 33 tokens per session scan A 29821fa68050
qa is a skill published in the GitHub repository 0xherve/qa-skill (2 stars, last pushed 2mo ago), licensed MIT. It adds 33 tokens to every session and 1,152 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
tutti-record-agent-session-replay
From a Tutti checkout, run, audit, freshly replay, publish, or diagnose Session Replay cassettes that are driven by case-repository scenario scripts (CDP), not by interactive UI recording. Use for real-Provider capture while a scenario.mjs executes, cassette transport or semantic-state mismatches, fresh replay…
testing-strategy
Selects the narrowest sufficient test boundary from requirements, repository evidence, and maintenance cost. Use when deciding integration or E2E coverage.
testing-fundamentals
Guides test strategy for unit, integration, and E2E testing. Use when junior asks "how do I test", "write tests", "what should I test", "test coverage", "mocking", or works with Vitest, Jest, Playwright. Provides testing pyramid and AAA patterns.
test-automation
Plan, write, and review automated tests following KATA (Komponent Action Test Architecture) on Playwright + TypeScript, or explain existing automated tests in a sealed read-only mode. Use when writing E2E or API/integration tests, creating Page or Api components, designing ATCs, parameterizing test data, registering…
integration-e2e-testing
Applies integration and E2E implementation, mock-boundary, and review rules. Use when: writing integration or E2E tests or reviewing their quality.
karpathy-guidelines
Behavioral guardrails for LLM-assisted coding. Use when writing, reviewing, or refactoring code in any project to avoid overcomplication, keep changes surgical, surface assumptions early, and execute against verifiable success criteria.