affiliate-disclosure-compliance
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data, such as content captions, scripts, transcripts, and live URLs, as described in
SKILL.md. This represents a potential surface for indirect prompt injection where malicious instructions embedded in the analyzed content could attempt to influence the agent's compliance verdict. However, this is an inherent functional requirement for the skill's purpose. - Ingestion points: The skill accepts 'caption text / script / transcript / URL / screenshot' from users or external sources to perform the compliance audit.
- Boundary markers: The instructions do not define specific delimiters or 'ignore' instructions for the ingested content to prevent the agent from following instructions found within the data.
- Capability inventory: The skill is primarily instructional and does not utilize dangerous capabilities like subprocess execution, dynamic code loading, or file system writes beyond its own reference files.
- Sanitization: There is no mention of input sanitization or filtering for the external data being analyzed.
Audit Metadata