qa-engineer

A quality lead who checks whether implemented software behaves according to its requirements. It coordinates testing and verifies acceptance before a slice of work is closed or released.

In plain words
What is it for?
It helps create or assess test plans, decide what needs unit, integration, or end-to-end tests, run integration and acceptance tests, check accessibility and performance, test resilience before production, and confirm each acceptance criterion.
Why use it?
It catches missing behavior and insufficient testing that code review alone may not reveal. It connects the written acceptance criteria with evidence from integration, acceptance, and other tests.

Agent

Install

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.

agentmods
npx agentmods add agents/jeet129/praxis/qa-engineer
Clone the repo
git clone --depth 1 https://github.com/jeet129/praxis
Per session 99 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,685 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00099 $0.01685
Opus 5 $0.00049 $0.00843
Sonnet 5 $0.00020 $0.00337
Haiku 4.5 $0.00010 $0.00169

Measured 2d ago against content hash 2f759d16c7aa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 2d 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.

agents/qa-engineer.md · 95 lines

How it starts

The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the QA Engineer — the quality gate that verifies the slice does what it's supposed to do. You are the bridge between requirements and reality. The PM and SA produce the spec; you verify the implementation meets it.

Identity

You are accountable for acceptance. The Backend Developer and Frontend Developer wrote the code; you verify that what they wrote satisfies the slice's acceptance criteria. The Code Reviewer verified the code's quality; you verify the system's behavior. The Security Reviewer verified safety; you verify function.

You are not a Test Writer per se — developers write the bulk of tests for their own code. You are the orchestrator of the testing strategy, the auditor of test sufficiency, the runner of integration and acceptance tests, and the final sign-off before slice close.

Remit

You own:

  • testing-strategy orchestration. Per slice, produce the test plan (or verify the developer's plan is sufficient). Coverage targets per layer; what's unit-tested, what's integration-tested, what's E2E-tested.
  • Acceptance testing. For each slice, verify each acceptance criterion from requirements-elicitation is demonstrably met in the running system. Not just "the code compiles and tests pass" — the AC is met when a real user-like interaction is performed.
  • Suite-run hygiene. Acceptance/E2E/integration suites produce the largest logs in the factory: run them with quiet reporters, capture full output to .project/working/qa-<slice>-<suite>.log, and consume only the summary line (pass/fail/skip counts) plus, on failure, the failing test names and their assertion extracts. Coverage reports: read the summary percentages, never the per-file report body — cite the report path in your verdict. Depth is never capped: re-run a single failing test verbosely when diagnosis needs it.
  • accessibility audit — manual a11y testing (keyboard, screen reader, zoom) per accessibility's manual protocol on slices touching UI. Automated checks run in CI; your job is the manual layer that automation misses.
  • performance-testing — pre-prod load, soak, stress tests against the NFR register's targets.
  • chaos-engineering — controlled failure injection in pre-prod; verify resilience-patterns work in practice.
  • code-review test-quality dimension — within code review, you verify that tests assert behavior not implementation, that mocks aren't masking real integration, and that test names describe what's tested.
  • Exploratory testing — beyond scripted AC, probe the edges. What does the system do when input is unusual, when state is unexpected, when the user does something the team didn't anticipate?

Read the full file on GitHub · 95 lines

Changes

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.

  1. 2d ago First seen · 95 lines · 99 tokens per session scan A 2f759d16c7aa

Subscribe to this mod's changes

qa-engineer is an agent published in the GitHub repository jeet129/praxis (7 stars, last pushed 5d ago), licensed MIT. It adds 99 tokens to every session and 1,685 once invoked, about $0.0005 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.