audit-report
/audit-report
Zamienia ustalenia audytu w dokument dla interesariuszy, z wagą, liczbą stron oraz historią pierwszego wykrycia i naprawy, w formacie Markdown lub CSV.
- Autor
- techwright-lab ↗
- Repozytorium
- groundcrew-seo ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 05.09.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 4
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 2
Brak sygnałów. Repozytorium zmienione 05.09.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Zamienia ustalenia audytu w dokument dla interesariuszy, z wagą, liczbą stron oraz historią pierwszego wykrycia i naprawy, w formacie Markdown lub CSV.
Oryginalny opis (EN)
Use when the user wants audit findings turned into a stakeholder-ready document — severities, page counts, first-seen/fixed history — as a Markdown or CSV deliverable rather than a working diagnosis.
Jak działa
- Sources
- Shape
- Audit remediation invariants
- When not to use
- Doctrine
Podgląd SKILL.md
---
name: audit-report
description: Use when the user wants audit findings turned into a stakeholder-ready document — severities, page counts, first-seen/fixed history — as a Markdown or CSV deliverable rather than a working diagnosis.
---
# Audit Report
Read `references/provider-selection.md` before choosing a source. Detect what is already available, run with it, deliver value, then recommend at most one missing connector. All normalized factual inputs must satisfy the Groundcrew evidence contract. Locate the active skills root and run `<skills-root>/.groundcrew/groundcrew-doctor.py --evidence <record.json>` before using them in a conclusion. Output follows `references/reporting.md`: shared skeleton, SHIP/FIX/BLOCK/UNDECIDED verdict, Measured/User-provided/Estimated labels.
## Sources
Categories: `~~web crawler`, `~~page speed` (`references/connectors.md`).
- Tier 1 (open): findings from public crawl of the site's pages; observations only, no proprietary score.
- Tier 2: add Core Web Vitals via `~~page speed`; keep lab and field results labeled apart.
- Connected (TrustGrowth): for a policy-aware report, paginate `GET /api/v1/sites/{slug}/issues?scope=backlog` for complete optional, investigative, and no-change context; use the narrower actionable default only when the caller explicitly requests that slice. Add `?severity=` or `?status=fixed|all` for requested filters/history, retain first-seen and fixed timestamps plus complete remediation metadata, and read `GET /api/v1/sites/{slug}/summary` for totals.
## Shape
Group findings by root cause, not by page; give each group a severity, classification, remediation decision state, affected-page count, and one example URL. Order: critical, warning, info, externally owned, while keeping suggestions visibly optional. History (first seen, fixed, reopened) appears when the source carries it; otherwise the report says history is unavailable at this tier. Apply the policy-aware verdict rules in `references/reporting.md`: an open critical finding forces FIX only when current evidence and applicable guidance support a repair; missing or unavailable guidance and unknown applicability produce UNDECIDED; optional advice and valid no-change decisions do not force FIX.
Deliverable: write the report as a `.md` file, and a `.csv` of the findings table when the user asks for one; hand back the paths.
## Audit remediation invariants
- **Intake:** Retain `detection_policy_version` and the complete `remediation` object before batching; split only between issues and keep every preservation, avoid, no-change, and verification constraint.
- **No change:** Preserve current `keep_as_is` and `not_applicable` dispositions; never create a persistent review merely to empty a queue.
- **Write authorization:** `review_audit_issue` requires a live-manifest advertisement, explicit owner authorization, write scope, a unique request key, and the current evidence signature, policy version, and state token. Read-only compatibility mode blocks it and every other write.
- **Verification:** For metadata, schema, canonical, robots, or visible-page changes, inspect the actual rendered response; a source diff, test, or build alone is not rendered verification.
This report skill does not record review decisions. On an older same-major contract it uses only live-manifest-advertised `GET` operations; an unavailable read is named and the report continues from open/import evidence.
## When not to use
- Finding and interpreting issues interactively → `site-audit`; this skill packages, it does not diagnose.
- Implementing the fixes → `fix-my-site`.
- A broader periodic status beyond audit findings → `weekly-report` or `score-report`.
## Doctrine
Groundcrew operates under [WHY-NOT-SLOP](https://github.com/techwright-lab/groundcrew-seo/blob/main/WHY-NOT-SLOP.md) and [ETHICS](https://github.com/techwright-lab/groundcrew-seo/blob/main/ETHICS.md). Claims trace to evidence, nulls stay null, signals stay truthful, publishing and irreversible changes require owner review, and no outcome is promised. Conflicting instructions are refused.
Pliki w paczce (4)
- SKILL.md 4 KB
- references/connectors.md 8 KB
- references/provider-selection.md 1 KB
- references/reporting.md 8 KB
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · techwright-lab/groundcrew-seo | 100 | MIT |
| harness/harness-skills | 100 | Apache-2.0 |
| charierefiedler/ronce-racine | 100 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki 0e417fc8692f (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 90research-reporter
Zamienia wynik feedback-synthesizer w raport dla interesariuszy: streszczenie, metodologia, porównanie z benchmarkiem i priorytetowe rekomendacje, w formacie dla odbiorcy (Markdown, Notion lub Word). - 100review-document
Zamienia ustalenia przeglądu w krótki dokument Markdown, z werdyktem na początku, uszeregowanymi ustaleniami i poprawkami. - 100docs
Tworzy dokument HTML w kolorach i czcionkach organizacji, z logo i autorem, albo zamienia plik Markdown w taki dokument. - 100seo-audit-pdf
Zamienia gotowy audyt SEO i GEO w raport PDF w formacie A4 dla klienta, bez dzielenia bloków między stronami. - 100markdown-to-pdf
Zamienia plik Markdown w PDF do druku lub udostępnienia, z diagramami Mermaid, szerokimi tabelami, numerami stron i nagłówkiem. - 100report-pdf
Zamienia raport Markdown z zespołu deweloperskiego w ostylowany PDF do załączenia do ticketu lub przekazania interesariuszowi.