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-reportgit 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-report)<a href="https://agentmods.dev/skills/pwdev-solucoes/pwdev-claude-marketplace/qa-report"><img src="https://agentmods.dev/badge/skills/pwdev-solucoes/pwdev-claude-marketplace/qa-report/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-report"><img src="https://agentmods.dev/badge/skills/pwdev-solucoes/pwdev-claude-marketplace/qa-report.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 | $0.00020 | $0.01399 |
| Opus 5.5 | $0.00008 | $0.00560 |
| Sonnet 5.5 | $0.00004 | $0.00280 |
| Haiku 4.5 | $0.00002 | $0.00140 |
Grade A, and why
qa-report 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Export a QA report
Export an already recorded QA run. This workflow publishes a report package; it does not create test results or reinterpret evidence commands. Read workflow, safety, artifacts, evidence, and reports first.
Inputs
- An explicit private manifest path and an explicit project root path. Never infer either path from the current directory, a run ID, or a prior conversation.
- The explicit objective or intent for the export, target identity, contract source/hash, complete acceptance-criterion catalog, cases, evidence, defects, limitations, recorded authorization, and existing QA verdict carried by that manifest.
- Explicit authorization to create a new report package beneath the supplied project root. The manifest's run ID must identify a new destination; export never grants any other authority.
Procedure
- Preserve the explicit objective or intent, target, contract, criteria, cases, evidence, defects, limitations, authorization, and recorded results. Do not infer missing criteria, applicability, assessments, execution, approval, or risk acceptance.
- Identify the affected QA surfaces and consult installed applicable
qa-specialist-requirements,qa-specialist-defects, or otherqa-specialist-*skills only to expose semantic gaps already present in the supplied record. An unavailable applicable specialist is a limitation. Specialist guidance does not expand authorization or change the normalized manifest. - Before export, require a complete acceptance criteria catalog that preserves all criterion
IDs, text, applicability and reason, assessment expected and observed values, and referenced
case IDs. Confirm the identified contract and completed criteria review are recorded. Missing
semantic confirmation remains a limitation and prevents a newly inferred
PASS. - Use the explicit manifest and project root unchanged. The workflow invokes the exporter
exactly once and never permits a second exporter invocation. Its only process invocation is
qa_report.py report --manifest PATH --project-root PATH. Replace eachPATHwith its separately supplied path; do not add another subcommand, omit either option, or reuse an implicit root. - Let the exporter enforce schema and verdict precedence. Before HTML or PDF rendering, the exporter must validate acceptance criteria, confinement, limits, hashes, target/contract binding, sanitization, and evidence safety. The exporter renders HTML and PDF from the same normalized manifest. It publishes only a complete validated package. Never edit the private or public manifest around validation.
- Preserve returned diagnostics, sanitization outcomes, and verdict without suppressing,
repairing, or reclassifying them. Preserve
export_status,output_dir, publication digest, and snapshot when present. Exit code0means both formats exported; it does not mean the QA verdict isPASS. Exit codes2and3remain refusal and incomplete export respectively. - Stored
commandvalues are inert text. It never executes or re-executes tests, manifest commands, or evidence commands. It never uses a stored command to fill a missing result and never retries the export after a failure or collision.
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 · 105 lines · 20 tokens per session scan A c8e77c57e520
qa-report 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 20 tokens to every session and 1,399 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
foundry-hosted-agent-validation
Step-by-step process for validating a Python Foundry hosted agent sample (under python/samples/04-hosting/foundry-hosted-agents/) end to end — running it locally (native runtime and azd ai agent run) and after deploying it to an Azure AI Foundry project with azd. Use this when asked to validate a hosted agent sample.
document-query
Read, extract or compare PDFs, Office files, URLs and large text; OCR when vision fails.
pdf-scan-config
Internal helper: manage .a11y-pdf-config.json scan settings.
kb-report-generation
Reference data, not a reviewer. Format accessibility audit reports with severity scoring (0-100, A-F grades), scorecard computation, and compliance exports including VPAT/ACR and remediation priorities.
kb-severity-mapping
Reference data, not a reviewer. Canonical severity level definitions and cross-domain mapping for web, document, and markdown audits. Score impact ranges, WCAG conformance alignment, and cross-format normalization.
yellow-sheet-corpus
Run the Yellow Sheet chain end to end against ANY matter, from the original PDFs, and read the result honestly. Use whenever someone asks what has been tested end to end, whether a matter produces a usable sheet, how to onboard a NEW matter, or asks to re-run after a code change. Carries the intake mechanism that…