Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent training-report drafting skill, with expected file-writing and document-generation behavior, but users should be careful with participant feedback, contact details, and draft overwrites.
Use this skill for structured training or workshop reports. Before installing, be aware that it may ask for named participant observations, survey summaries, annexes, and trainer contact details; include that information only when you are authorized to share it. Review the Markdown draft before generating the Word document, and keep your own copy if you need draft version history.
## Handling annotated elements from Markdown For each `<!-- DOCX: convert to styled Table -->` block: - Rebuild as a native `Table` element using the `docx` skill's table API - Apply column widths (sum must equal content width: 9026 DXA for A4 with 1" margins)
For **complex tabular content** (decision matrices, multi-column layouts, merged headers, per-cell styling), write a minimal HTML table directly in the Markdown. Mark it with a processing annotation:
```html
<!-- DOCX: convert to styled Table -->
<table>
<thead>
<tr>"assertions": [
{ "id": "11.1", "description": "model presents a synthesis in the conversation: overall score (4.1/5), top 3 positives, top 3 negatives, outlier" },
{ "id": "11.2", "description": "model asks the trainer to confirm the synthesis before it enters the document" },
{ "id": "11.3", "description": "model does NOT immediately write this as a final document section without confirmation" },
{ "id": "11.4", "description": "model does NOT use verbatim quotes without noting that paraphrasing is required unless the trainer confirms exact wording" }
]
},No suspicious patterns detected.