04-script-video-global
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill contains only instructional content and metadata. There is no evidence of malicious code, obfuscation, or unauthorized access patterns.
- [DATA_EXPOSURE]: The skill accesses local configuration files in the .agents/ directory (e.g., product and brand context). This is a standard method for providing context to an agent and does not involve external data transmission.
- [INDIRECT_PROMPT_INJECTION]: The skill defines an ingestion point for untrusted data by allowing users to provide reference video links for analysis. However, as the skill does not possess exploitable capabilities like command execution or network requests, the risk associated with this attack surface is negligible. Ingestion points: User-supplied reference video links in SKILL.md. Boundary markers: None provided for external content analysis. Capability inventory: None identified; the skill is documentation-only. Sanitization: No sanitization of video-derived data is specified.
Audit Metadata