notion-doc
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a provided Python script (
lint.py) locally to validate that the generated HTML documents adhere to the style guide. - [EXTERNAL_DOWNLOADS]: The
template.htmlfile includes references to Prism.js libraries hosted on Cloudflare's CDN (cdnjs.cloudflare.com) to provide code syntax highlighting. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided content to generate styled documents, creating a surface where malicious instructions embedded in the data could influence the agent's behavior during document creation.
- Ingestion points: User-supplied content for documents, reports, or summaries processed in
SKILL.md. - Boundary markers: None explicitly defined to isolate user content from the document's design instructions.
- Capability inventory: Execution of shell commands (
python3) and file system access (writing and reading HTML/Markdown files). - Sanitization: No specific sanitization or escaping of external content is documented before interpolation into the Notion-style template.
Audit Metadata