baoyu-infographic
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user content and interpolates it into a prompt for an image generation agent. Ingestion point: SKILL.md (Workflow Step 1.2). Capability inventory: Calls external image generation skill (SKILL.md Step 6). Boundary markers: None explicitly mentioned in the interpolation template references/base-prompt.md. Sanitization: Instructions emphasize verbatim copying but lack specific sanitization for prompt-breaking characters.
- [COMMAND_EXECUTION]: The skill uses basic shell commands (test -f) to check for the existence of its own configuration files in the project and home directories. This is standard configuration loading behavior.
Audit Metadata