image-convert
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
agent-mediacommand-line utility for its core functionality. Evidence: The commandagent-media image convertis defined in the skill markdown. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection. Ingestion points: The
--inparameter accepts data from local file paths and remote URLs. Boundary markers: The skill does not define specific delimiters to separate image data from other agent instructions. Capability inventory: Theagent-mediatool can perform file system operations and fetch data from the network. Sanitization: There is no evidence of sanitization for image metadata or content which may contain embedded instructions.
Audit Metadata