listenhere
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill acts as a comprehensive API manual for the ListenHere service. It defines endpoints, expected request/reponse formats, and rate limiting policies. There are no malicious instructions, hardcoded credentials, or unauthorized command execution patterns detected.
- [INDIRECT_PROMPT_INJECTION]: The skill describes functionality for reading data from external, user-controlled sources which could theoretically contain instructions intended to influence the agent's behavior.
- Ingestion points: Public comments retrieved from the
/v1/releases/{release_id}/commentsand/v1/tracks/{track_id}/commentsendpoints, as well as the contents of synchronized lyric JSON files. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when processing fetched comments or lyrics.
- Capability inventory: The skill grants the agent the ability to manage accounts, upload files to R2 storage, publish music releases, and perform moderation actions (e.g., hiding comments or suspending accounts).
- Sanitization: No specific client-side sanitization or validation of the text content within comments or lyrics is mentioned for the reading workflow.
Audit Metadata