favicon
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests a user-provided file path and processes the content of that file (including SVGs), which represents an attack surface for indirect prompt injection. \n
- Ingestion points: The source image path ($1) and the source image file content are ingested by the agent as described in SKILL.md. \n
- Boundary markers: No explicit boundary markers are used to isolate untrusted file paths or content from the agent's internal instructions. \n
- Capability inventory: The skill utilizes shell execution (magick, cp, mkdir) and has file write access to the local project structure. \n
- Sanitization: Instructions suggest checking for file existence and extension, but do not provide specific guidance for sanitizing paths or escaping shell metacharacters before execution.\n- [COMMAND_EXECUTION]: The skill executes shell commands by interpolating user-provided input ($1) into command strings. This pattern is susceptible to argument injection if the input path is maliciously formatted, even though the tools are restricted by the allowed-tools configuration.
Audit Metadata