html-report-builder
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data (e.g., spreadsheets, API exports) provided by the user to generate report content, which creates a potential surface for indirect prompt injection. 1. Ingestion points: User-provided data identified in SKILL.md. 2. Boundary markers: Absent; no specific delimitation instructions are provided to the agent. 3. Capability inventory: No command execution or network-based scripts are defined within the skill. 4. Sanitization: No data validation or escaping protocols are specified for the input data.
- [EXTERNAL_DOWNLOADS]: The CSS framework in REFERENCE.md fetches font stylesheets from Fontshare's CDN (api.fontshare.com), which is a common and legitimate practice for web documents.
Audit Metadata