content-asset-creator

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes curl commands to transmit data to the Gamma Generate API and the v0 Platform API. It also suggests using npx playwright for converting the generated HTML into PDF format. These commands are standard for API interaction and document processing in this context.
  • [EXTERNAL_DOWNLOADS]: Styling and layout resources are fetched from well-known content delivery networks (CDNs), specifically Tailwind CSS and Google Fonts. These are trusted infrastructure services used for web asset styling.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes untrusted structured data to populate its templates. Ingestion points: Title, narrative, and stat fields defined in the input structure of SKILL.md. Boundary markers: Absent. Capability inventory: File system write operations for HTML files and subprocess execution via playwright in SKILL.md. Sanitization: Not explicitly mentioned for user-supplied data. This allows for potential injection of malicious text or script content into the final generated asset.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:05 PM
Security Audit — agent-trust-hub — content-asset-creator