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 skills add arozumenko/sdlc-skills --skill quality-evalsgit clone --depth 1 https://github.com/arozumenko/sdlc-skillsWrote 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/arozumenko/sdlc-skills/quality-evals)<a href="https://agentmods.dev/skills/arozumenko/sdlc-skills/quality-evals"><img src="https://agentmods.dev/badge/skills/arozumenko/sdlc-skills/quality-evals/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/skills/arozumenko/sdlc-skills/quality-evals"><img src="https://agentmods.dev/badge/skills/arozumenko/sdlc-skills/quality-evals.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 | $0.00109 | $0.00934 |
| Opus 5.5 | $0.00044 | $0.00374 |
| Sonnet 5.5 | $0.00022 | $0.00187 |
| Haiku 4.5 | $0.00011 | $0.00093 |
Grade A, and why
quality-evals 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
quality-evals
Two of this bundle's agents make judgment calls that are easy to trust and
hard to verify: test-runner decides PASS/FAIL/BLOCKED on your live app,
and test-author decides how a rough idea becomes a formatted TC-NNN.md.
Trusting either without measurement is a leap of faith. This skill ports a
rigorous, held-out-answer-key evaluation methodology — built and proven over
months against several real apps — so any team can measure their own
installation's accuracy against their own app, not just take someone
else's word for it.
When to use
- Before rolling the manual-qa bundle out to a wider team or a release-gating workflow, and you want a number, not a feeling.
- After a
test-author/RULES.mdortest-runner/AGENT.mdchange (yours or upstream's), to confirm accuracy didn't regress. - When comparing model choices, prompt tweaks, or a fork of these agents against a stable baseline on your own suite.
- When you want to build a small gold suite for your own app: a set of test cases with a known-correct answer, reusable for regression-testing the agents themselves going forward.
What it is NOT
Not a replacement for test-reporter's run reports, and not something that
runs automatically as part of a normal test-run-lead session. This is a
deliberate, opt-in benchmarking exercise you run when you want to measure
the agents — same relationship the tokenomics skill (in the
test-automation bundle) has to ordinary automation work: dormant until you
invoke it.
Two tracks
| Track | Measures | Agent under test | Scored by |
|---|---|---|---|
| Detection | Does test-runner correctly report PASS/FAIL against a known-correct answer, including catching a bug it was never told about |
test-runner |
scripts/score-test-runner-output.mjs (fully deterministic) |
| Authoring | Does test-author turn a rough input into a structurally correct, semantically faithful test case |
test-author |
scripts/score-test-author-output.mjs (Tier A) + references/judge-rubric.md (Tier B) |
What ships with it
6 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.
- 1mo ago First seen · 75 lines · 109 tokens per session scan A 4c12437f76b8
quality-evals is a skill published in the GitHub repository arozumenko/sdlc-skills (25 stars, last pushed yesterday), licensed MIT. It adds 109 tokens to every session and 934 once invoked, about $0.0004 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 2026-09-13.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.