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 sumithr/sumo-qa --skill sumo-qa-finishing-qa-workgit clone --depth 1 https://github.com/sumithr/sumo-qaWrote 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/sumithr/sumo-qa/sumo-qa-finishing-qa-work)<a href="https://agentmods.dev/skills/sumithr/sumo-qa/sumo-qa-finishing-qa-work"><img src="https://agentmods.dev/badge/skills/sumithr/sumo-qa/sumo-qa-finishing-qa-work/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/sumithr/sumo-qa/sumo-qa-finishing-qa-work"><img src="https://agentmods.dev/badge/skills/sumithr/sumo-qa/sumo-qa-finishing-qa-work.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 107 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00102 | $0.02303 |
| Opus 5 | $0.00051 | $0.01151 |
| Sonnet 5 | $0.00020 | $0.00461 |
| Haiku 4.5 | $0.00010 | $0.00230 |
Grade A, and why
sumo-qa-finishing-qa-work 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 10d 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Finishing QA work
Close the loop on a QA rollout. The execution is done; now the user needs evidence they can show their team / their reviewer / their PR.
Announce at start: "Capturing evidence and writing the summary."
Output discipline (mandatory)
Inherits the global discipline from using-sumo-qa: output discipline (never surface internal taxonomy labels — say "behaviour change in pricing", not "Classification: business_logic_change"), output economy (spend output on findings not framing; no preamble or self-narration; one question per turn; no closing pleasantries), knowledge authority hierarchy, internal scaffolding stays internal, and specialty-tool fit.
Closing-line integrity: the closing recap must agree with the body. If the body lists N known gaps, the closing recap must say "N known gaps", not "0 known gaps". No closing confirmation question — the summary IS the record; the user can ask follow-ups in their own time.
The Iron Law
NO FINISH WITHOUT FRESH EVIDENCE + WRITTEN SUMMARY. The deliverable is a written summary another engineer could read tomorrow and know what was covered, what wasn't, and what to do next.
Evidence-backed gate reporting
Every claim in the summary — "suite green", each risk covered, a known gap, "ready to release" — is a gate claim with a status (passed / failed / skipped / blocked / unverified) that, unless skipped or unverified, cites the ONE observed evidence item behind it by source (command, tool_call, file_read, user_fact, external_ci, manual_observation). A passed / failed / blocked claim with no cited source overstates; unverified is the honest state when nothing was observed this turn (a KNOWN GAP is unverified or failed, never a silent passed). Keep it compact — a status word + short source cite per line, not a second narrative.
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.
- 10d ago First seen · 129 lines · 102 tokens per session scan A 9892591c33e6
sumo-qa-finishing-qa-work is a skill published in the GitHub repository sumithr/sumo-qa (6 stars, last pushed today), licensed Apache-2.0. It adds 102 tokens to every session and 2,303 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.
Other skills, from other repositories
live-audit
Audit steam-games-mcp — build/test/lint gate, live MCP tool edge-case sweep (input validation, SteamID64/vanity/appid edge cases, key-gating), and source-level code review. Use when asked to test/audit the published or just-fixed steam-games-mcp package, hunt for bugs/edge cases, or repeat "the same kind of testing as…
pr-quality-cycle
Combined PR quality workflow: address inline review comments AND fix Codecov patch coverage gaps in one end-to-end cycle. Use when a PR has both review feedback and coverage regression, or when you want a single checklist that covers triage, code fixes, test additions, verification, and push-gate validation before…
qa
QA test a live website with Axon discovery/content evidence plus browser automation when interaction is required. Use when the user wants exploratory QA, form testing, navigation/link checks, responsive checks, performance observations, bug reports, or a pre-launch quality review.
adversarial-reviewer
Adversarial code review that assumes bugs exist and hunts for them. Use when asked to review code, find bugs, audit for correctness, stress-test a PR, or when someone says "tear this apart" or "what's wrong with this". Give no benefit of the doubt — every line is guilty until proven innocent.
brooks-sweep
Full-sweep mode: runs a unified analysis across all quality dimensions — code decay, architecture, tech debt, and test quality — then applies fixes directly to the codebase. Safe changes are auto-applied; risky changes are confirmed before execution. Drawing on twelve classic engineering books. Triggers when: user…
harness-engineering-lifecycle
Design, implement, and integrate generalized validation harnesses across a producer-consumer boundary after a local harness contract exists. Use when refactoring custom validation CLIs/MCPs for large polyrepos, extending Steward across sibling repos, or deploying a local tool to a consumer project for dogfooding and…