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 pwdev-solucoes/pwdev-claude-marketplace --skill qa-buggit clone --depth 1 https://github.com/pwdev-solucoes/pwdev-claude-marketplaceWrote 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/pwdev-solucoes/pwdev-claude-marketplace/qa-bug)<a href="https://agentmods.dev/skills/pwdev-solucoes/pwdev-claude-marketplace/qa-bug"><img src="https://agentmods.dev/badge/skills/pwdev-solucoes/pwdev-claude-marketplace/qa-bug/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/pwdev-solucoes/pwdev-claude-marketplace/qa-bug"><img src="https://agentmods.dev/badge/skills/pwdev-solucoes/pwdev-claude-marketplace/qa-bug.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 | $0.00024 | $0.01757 |
| Opus 5.5 | $0.00010 | $0.00703 |
| Sonnet 5.5 | $0.00005 | $0.00351 |
| Haiku 4.5 | $0.00002 | $0.00176 |
Grade A, and why
qa-bug 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Record and reassess a defect
Record defect facts without changing the product. Preserve reproduction, triage, scope, evidence, attempt history, and retest semantics under the governing contract. Read workflow, safety, artifacts, and evidence first.
Inputs
- Project root, target/environment/build/change identity, explicit objective or intent, identified contract source/hash, and the complete supplied criterion catalog with exact IDs, text, applicability, and current results.
- Stable defect ID, summary, prerequisites, exact reproduction steps and data, frequency, expected and observed behavior, affected surface, in-scope/out-of-scope decision, and local target-bound evidence.
- Severity and priority proposals with separate rationales, owner, current state, linked criterion IDs or an explicit unlinked reason, related risks, and complete current-defect inventory.
- Stable logical case ID, original failure and all retest attempts with unique attempt IDs, increasing attempt numbers, supersession links, evidence, and explicit authorization for this write operation or any separately requested retest execution.
Procedure
- Preserve the explicit objective or intent, target, environment/build, contract, criterion IDs/text, defect identity, existing attempts, approvals, and authorization exactly. Do not infer expected behavior, applicability, resolution, risk acceptance, or execution permission.
- Identify affected surfaces and consult the installed applicable
qa-specialist-*skills, starting withqa-specialist-defects; consult requirements, functional, API, Web, mobile, data, security, performance, automation, CI/CD, regression, or production specialists only when their surfaces apply. An unavailable applicable specialist is a limitation. Specialist guidance cannot expand authorization. - Materialize reproduction with prerequisites, ordered steps/actions, safe data, environment and
build, frequency, expected oracle, observed mismatch, scope and rationale, criterion links or
explicit unlinked reason, risk links, and verified evidence references. Insufficient
reproduction or evidence is
BLOCKED; never manufacture certainty. - Assess severity from product/user impact and blast radius. Assess priority independently from delivery order, urgency, dependencies, and scheduling. Record both rationales, owners, and scope; neither value rewrites the other, the defect status, or the QA result.
- Preserve the original failure and every retest as immutable history. Use one stable logical
case ID, unique attempt IDs, strictly increasing attempt numbers, the same target, and a
linear
supersedeschain. Never delete a failure, branch/cycle history, or rerun until green. - Record a retest result only from an explicitly authorized
qa-testexecution and valid local, confined, target/contract-bound, integrity-checked, sanitized evidence. This write workflow does not itself execute a stored command or product correction. - Mark this defect resolved only when the referenced terminal retest attempt is
PASSwith valid current evidence; a blocked/failed terminal attempt never falls back to an older pass. Return globalPASSonly after the terminal retest is valid and has valid current evidence, all applicable criteria arePASS, no pending work or limitations remain, and no other current in-scope defects remain. Missing criteria or defect inventory isBLOCKED; another proven current in-scope defect isFAIL. - Apply verdict precedence: any proven current in-scope failure, including an unlinked defect,
is
FAIL; without such proof, missing reproduction, evidence, retest, or complete inventories isBLOCKED.qa_report.py report --manifest PATH --project-root PATHis a later inert export and the report does not execute or re-run tests.
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 · 124 lines · 24 tokens per session scan A ccaf5b550cb8
qa-bug is a skill published in the GitHub repository pwdev-solucoes/pwdev-claude-marketplace (3 stars, last pushed 10d ago), licensed Apache-2.0. It adds 24 tokens to every session and 1,757 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 2026-09-16.
Other skills, from other repositories
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
langsmith-observability
LLM observability platform for tracing, evaluation, and monitoring. Use when debugging LLM applications, evaluating model outputs against datasets, monitoring production systems, or building systematic testing pipelines for AI applications.
experimental-code-coverage-local-debugger
Runs code coverage locally via Universal Test Runner (UTR) or helper scripts, mimicking LUCI trybots. Activate when CQ tryjobs fail or underreport coverage, to test local GN/recipe repairs before uploading, or to debug hermetic crashes.
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
cli-e2e
Write, modify, or debug Docker-based Composio CLI end-to-end tests under ts/e2e-tests/cli, including binary invocation, fixture isolation, output assertions, and package manifests. Use for CLI E2E test suites only; use cli-command for CLI source implementation.
adversarial-reviewer
Stress-test a code change for concrete correctness defects, unsafe assumptions, and failure modes. Use when the user explicitly asks for an adversarial review, a deep bug hunt, or to tear a change apart, or to check your own change before submitting it. Do not use for ordinary review, style feedback, or general…