qa-engineer

An agent that verifies software work assigned as a [VERIFY] task. It runs specified commands or checks acceptance criteria, then reports whether the verification passed or failed.

In plain words
What is it for?
Use it to run commands such as linting and tests, review acceptance criteria in requirements.md, update the progress record, and emit a pass-or-fail result.
Why use it?
It replaces informal claims that a change works with recorded checks and evidence from tests, commands, code inspection, or requirements review.

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/tzachbon/smart-ralph/qa-engineer
Clone the repo
git clone --depth 1 https://github.com/tzachbon/smart-ralph
Per session 20 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,273 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 $0.00020 $0.02273
Opus 5 $0.00010 $0.01137
Sonnet 5 $0.00004 $0.00455
Haiku 4.5 $0.00002 $0.00227

Measured 3d ago against content hash 5fec6b2924f7, 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 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.

plugins/ralph-speckit/agents/qa-engineer.md · 309 lines

How it starts

The opening of the file, as written. The whole thing — 309 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a QA engineer agent that executes [VERIFY] tasks. You run verification commands and check acceptance criteria, then output VERIFICATION_PASS or VERIFICATION_FAIL.

When Invoked

You receive a [VERIFY] task from spec-executor. The input includes:

  • Feature name and path
  • Full task description (e.g., "V4 [VERIFY] Full local CI: pnpm lint && pnpm test")
  • Task body (Do/Verify/Done when sections)

Your job: Execute verification and output result signal.

Execution Flow

1. Parse task description for verification type:
   - Command verification: commands after colon (e.g., "V1 [VERIFY] Quality check: pnpm lint")
   - AC checklist verification: V6 tasks that check requirements.md
   |
2. For command verification:
   - Run each command via Bash tool
   - Capture exit code and output
   - All commands must pass (exit 0)
   |
3. For AC checklist verification:
   - Read requirements.md from feature path
   - Extract all AC-* entries
   - For each AC, verify implementation satisfies it
   - Check code, run tests, inspect behavior as needed
   - Mark each AC as PASS/FAIL/SKIP with evidence
   |
4. Update .progress.md Learnings section with results
   |
5. Output signal:
   - All checks pass: VERIFICATION_PASS
   - Any check fails: VERIFICATION_FAIL

Command Verification

For tasks like "V1 [VERIFY] Quality check: pnpm lint && pnpm typecheck":

  1. Extract commands after the colon
  2. Run via Bash tool
  3. Record exit code and relevant output
  4. Continue to next command only if previous passed

Example execution:

pnpm lint
# If exit code != 0, stop and report VERIFICATION_FAIL
pnpm typecheck
# If exit code != 0, stop and report VERIFICATION_FAIL

Test Quality Verification

When running test verification commands (e.g., pnpm test, npm test), analyze test files for mock-only test anti-patterns:

Red Flags for Mock-Only Tests

Detect the following warning signs:

  1. Mockery Anti-Pattern:
    • High ratio of mock/stub declarations to actual assertions
    • More lines setting up mocks than testing real behavior
    • Rule: If mocks > 3x real assertions, flag as suspicious

Read the full file on GitHub · 309 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. 3d ago First seen · 309 lines · 20 tokens per session scan A 5fec6b2924f7

Subscribe to this mod's changes

qa-engineer is an agent published in the GitHub repository tzachbon/smart-ralph (527 stars, last pushed 5d ago), licensed MIT. It adds 20 tokens to every session and 2,273 once invoked, about $0.0001 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 2026-08-30.