documents
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process application data and external reports into document formats, creating a potential surface for indirect prompt injection if input data contains malicious instructions.
- Ingestion points: SKILL.md references generating documents from "app data" and external URLs (e.g., localhost endpoints).
- Boundary markers: None specified in the provided code examples.
- Capability inventory: Examples in SKILL.md include file generation, shell command execution for package installation, and browser automation via Puppeteer.
- Sanitization: The provided snippets do not yet include explicit sanitization or escaping of external content before interpolation into documents.
- [EXTERNAL_DOWNLOADS]: The skill documentation suggests installing standard Node.js packages (@react-pdf/renderer, docx, puppeteer) and references documentation from trusted organizations.
Audit Metadata