cardiology-youtube-scriptwriter
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- No Code (SAFE): The skill consists of 5 Markdown files (SKILL.md and four reference files). There are no executable scripts, dependencies, or system-level configuration files that could trigger malicious actions.
- Indirect Prompt Injection (LOW): The skill includes a 'Social Listening' phase (Phase 2) that ingests data from untrusted external sources like Reddit and YouTube comments. This creates a surface for indirect prompt injection where malicious content in search results could attempt to influence the agent's behavior. However, the severity is low because the agent has no capabilities to perform file modifications, execute commands, or access sensitive data. 1. Ingestion points: Phase 2 in SKILL.md and references/social-listening.md. 2. Boundary markers: Absent. 3. Capability inventory: Limited to text generation for scriptwriting. 4. Sanitization: No sanitization of search results is defined.
Audit Metadata