docx
Create, inspect, edit, restyle, review, compare, sanitize, render, audit, and finalize professional Microsoft Word .docx documents. Use this skill whenever PilotDeck must produce or modify a Word document, preserve an existing document while making targeted changes, add comments or tracked replacements, analyze document structure or metadata, verify accessibility and layout quality, compare revisions, remove review data, or deliver a visually checked DOCX. Use only for .docx files, not legacy .doc, macro-enabled .docm, or Google Docs operations.
80
100%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Low
Low-risk findings worth noting
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
The required runtime path for DOCX ingestion is via `bash "$DOCX_SKILL_ROOT/scripts/docx.sh" inspect ...` → `scripts/docx_cli.py:85` → `scripts/docxlib/core.py:221-286` which extracts free text from the user-supplied DOCX (`paragraph.text`, headings, table cell text, comments, etc.); if that DOCX was authored/modified by an outsider, the skill will pass that outsider-authored prose through to downstream LLM context as “inspection JSON / extracted fields” (indirect prompt-injection risk).
- Repository
- OpenBMB/PilotDeck
- Commit
c88d7d1
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.