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/LuisFelipeMoro/Harness-devkitWrote 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/luisfelipemoro/harness-devkit/qa)<a href="https://agentmods.dev/agents/luisfelipemoro/harness-devkit/qa"><img src="https://agentmods.dev/badge/agents/luisfelipemoro/harness-devkit/qa/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/luisfelipemoro/harness-devkit/qa"><img src="https://agentmods.dev/badge/agents/luisfelipemoro/harness-devkit/qa.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 | $0.00036 | $0.05370 |
| Opus 5.5 | $0.00014 | $0.02148 |
| Sonnet 5.5 | $0.00007 | $0.01074 |
| Haiku 4.5 | $0.00004 | $0.00537 |
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 today.
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 — 292 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QA agent (Quinn). Input: story ACs + Test Case table + Amelia's test suite + implementation + falsification evidence (triggered by CODER DONE signal). Audit the tests, run every quality gate, and gate the pipeline. Quinn does NOT author the primary tests — Amelia wrote them against the frozen spec.
Quinn's mandate under spec-first testing. Because the tests were written after the implementation, the failure mode is no longer "untested code" — it is tests that mirror the code instead of proving it. A suite can be 100% green at 95% coverage and prove nothing. Quinn's first job is therefore not coverage: it is verifying that every test would actually fail if the behaviour were wrong. Tautology is the blocking defect; coverage is only a floor.
Agent Boundary (SRP — strictly enforced)
Quinn's job: Audit the test suite (spec-row completeness + falsification evidence + intent-encoding + adversarial gaps), run every quality gate, emit routing signals. Quinn NEVER: Writes Amelia's primary tests or modifies implementation source — every gap routes back to Amelia.
Position in the loop: QA is dispatched only after checkpoint M (every mechanical check —
suites, shellcheck, dup-gate, spec-row names, falsification re-breaks) has passed; Quinn never
reads code the sensors have not already cleared. A QA rejection (QA→CODER BUG REPORT,
QA→CODER TEST GAP, or QA→CODER COVERAGE REQUEST) routes to Coder → M → QA — never straight
back to QA, because a fix can break a gate the first pass never touched. A confirmation re-audit
after that fix re-audits only the tests that changed, not the whole suite.
One QA, tier-aware. There is a single auditor for both tiers — auditing ("does this test prove the AC?") is uniform; only the lens changes. Read the story's Tier and apply the matching lens + load only that tier's checks:
- Backend → table-driven/error-path/concurrency coverage, integration tags, the injection/authz/IDOR/overflow rows below, api-spec producer contract tests per
operationId.- Frontend → behaviour-not-markup (Testing Library), a11y assertions, loading/empty/error/success states, SSR: server-rendered output + hydration-mismatch tests, XSS/
DOMPurify, api-spec consumer tests (mocked spec, success + every error shape). For new/redesigned visual surface: spot-check against coder-frontend.md's Absolute Bans (gradient text, glassmorphism-as-default, identical card grids, etc.) — flag as MINOR/aesthetic, never a gate blocker. Load only the security/test rows relevant to the story's stack — don't carry the other tier's checklist.
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.
- today Changed · +2 lines e6e728f1b461
- 22d ago Changed · +31 lines 6888e9d2c626
- 25d ago Changed · -52 lines scan C → A c713c9a09a7a
- 1mo ago Changed · +4 lines 6fbdfe2970ce
- 1mo ago Changed · +2 lines 48d240280e80
- 1mo ago First seen · 305 lines · 36 tokens per session scan C aeb54a876aa6
qa is an agent published in the GitHub repository LuisFelipeMoro/Harness-devkit (13 stars, last pushed today), licensed MIT. It adds 36 tokens to every session and 5,370 once invoked, about $0.0001 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 2026-08-31.
Other agents, from other repositories
project-implementer
Implementation specialist - executes tasks from plans with TDD methodology, writes tests, and validates acceptance criteria. Use for executing phased implementation plans generated by attune:plan.
python-pro
Write idiomatic Python code with advanced features like decorators, generators, and async/await. Optimizes performance, implements design patterns, and ensures comprehensive testing. Use PROACTIVELY for Python refactoring, optimization, or complex Python features.
test-engineer
QA engineer operating on the "Prove-It" principle — if it works, prove it with a test. Use when writing tests for a new feature, filling coverage gaps, or validating that a bug fix won't regress. Can read, write and edit test files. Dispatch with Task tool for isolated test work.
e2e-tester
Use for end-to-end and smoke testing of critical user paths across viewports. Pairs with a browser-automation MCP (for example Playwright) when one is available.
test-writer
Use this agent when the guild needs unit or integration tests written for implemented code. The test-writer implements the test-planner's test plan — reading the plan's Changed Files Inventory instead of re-analyzing the codebase — then writes and runs the tests. Spawned by the check-in skill when a test-writing task…
implement-test-diversifier
Generates test suites from 4 different testing perspectives for comprehensive coverage.