sonos
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes metadata from the local Sonos network, such as track titles and playlist names, which could theoretically contain instructions intended to influence the agent.
- Ingestion points: Command outputs from
sonos status,sonos list, andsonos favorites(SKILL.md). - Boundary markers: The skill does not define specific delimiters or warnings for the agent to ignore instructions embedded in track metadata.
- Capability inventory: The skill uses
Bash,Read, andWritetools (SKILL.md frontmatter). - Sanitization: No explicit sanitization or filtering of the metadata retrieved from the speakers is implemented.
Audit Metadata