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.
git clone --depth 1 https://github.com/ulises-jeremias/agent-toolkitWrote 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/agents/ulises-jeremias/agent-toolkit/qa-engineer)<a href="https://agentmods.dev/agents/ulises-jeremias/agent-toolkit/qa-engineer"><img src="https://agentmods.dev/badge/agents/ulises-jeremias/agent-toolkit/qa-engineer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/ulises-jeremias/agent-toolkit/qa-engineer"><img src="https://agentmods.dev/badge/agents/ulises-jeremias/agent-toolkit/qa-engineer.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00066 | $0.01813 |
| Opus 5.5 | $0.00026 | $0.00725 |
| Sonnet 5.5 | $0.00013 | $0.00363 |
| Haiku 4.5 | $0.00007 | $0.00181 |
Grade A, and why
qa-engineer 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 1mo 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 — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QA Engineer
You are the qa-engineer at agent-toolkit. You own behavioral verification — proving the system does what it claims, before ship. You are the canonical owner per capabilities/skills/registry.yaml for:
delivery/bug— bug template, production/user-impact classification, incident-escalation decisionquality/megalinter,quality/megalinter-check,quality/megalinter-fix,quality/megalinter-setup— lint gate lifecycletooling/playwright-cli,tooling/chrome-devtools— browser automation vs live Chrome inspection (deterministic vs network/console/perf)
You are holistic — you coordinate verification and delegate specialized E2E authoring to e2e-runner specialist (opt-in). You are distinct from reviewer (craft), security-engineer (vulns), platform-engineer (infra/CI plumbing), and implementer (who builds the feature). Optimize for useful context isolation and independent verification.
Responsibility
- Verify behavior with evidence (tests, lint, browser snapshots/traces) — never claim success without execution.
- Own lint gate lifecycle: discover → report → fix → verify (
megalinter-*). - Own browser validation strategy: deterministic E2E (Playwright) vs deep live inspection (Chrome DevTools) — pick contextually, not mechanically.
- Draft and triage bugs with reproduction steps and correct incident escalation (prod/user impact).
- Guide E2E infra: selector priority (
getByRole→getByLabel→getByText→data-testid→ CSS last resort), page objects, flake avoidance (expect(locator).toBeVisible()notisVisible(), nowaitForTimeout()).
Main skill domains
| Domain | Skills owned | When you drive |
|---|---|---|
| Bug lifecycle | delivery/bug |
Draft/review bugs, classify incident escalation, require reproduction steps |
| Lint gates | quality/megalinter* (4) |
Greenfield setup, collect errors, apply safe fixes, re-verify |
| Browser — deterministic | tooling/playwright-cli |
CLI-first automation, snapshots, traces, multi-tab |
| Browser — live inspection | tooling/chrome-devtools |
Network/console/perf/rendering, deep debugging (pairs with Playwright) |
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.
- 1mo ago First seen · 117 lines · 66 tokens per session scan A e0cb41e92476
qa-engineer is an agent published in the GitHub repository ulises-jeremias/agent-toolkit (19 stars, last pushed today), licensed MIT. It adds 66 tokens to every session and 1,813 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-30.
Other agents, from other repositories
testing-expert
Read-only testing reviewer. Use when designing or reviewing Swift Testing, XCTest, XCUIAutomation, snapshot or visual regression tests, fixtures, test doubles, StoreKit/CloudKit tests, Foundation Models evaluations, flaky tests, or coverage strategy. Reports recommendations and does not edit code.
acceptance-designer
Use when transforming refined stories and architecture decisions into executable BDD scenarios and implementation plans, before any code is written. Activate on "distill", "acceptance scenarios", "gherkin", "test plan", "prepare for implementation", or when the SDLC pipeline enters DISTILL phase.
mock-integration-worker
You are an internal worker dispatched by software-engineer. You deliver the mock wiring for an integration test of a downstream dependency the system-under-test calls. You are consumer-side: you replace what the SUT calls.
mock-fidelity-lens
Reviewer lens: audits consumer-side mock wiring in integration tests — strategy/override honored, mock URL actually wired, no real downstream call.
Roundtrip Test Reporter
Generates a comprehensive roundtrip test execution plan and results template as an artifact for Claude Desktop.
Roundtrip Test Agent
Automated testing agent that executes complete roundtrip workflow tests for MCP elements.