teaching-html-story-deck
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the
npx postplancommand to upload generated HTML decks. This triggers a download and execution of thepostplanpackage from the public npm registry. - [COMMAND_EXECUTION]: Several shell commands are executed during the workflow:
python scripts/init_deck.pyis used for scaffolding,python scripts/validate_deck.pyis used for quality checks, andnpx postplanis used for deployment. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided content to generate HTML, which presents an indirect prompt injection surface. 1. Ingestion points: Raw content and source material provided by the user in
SKILL.md. 2. Boundary markers: The instructions lack explicit delimiters or 'ignore embedded instructions' warnings for the input data. 3. Capability inventory: The skill can write files and perform network operations vianpx postplanand handoffs tohtml-stable-publish. 4. Sanitization: Thescripts/validate_deck.pyscript performs a post-generation check to ensure no unauthorized external scripts or styles were added to the output.
Audit Metadata