baoyu-comic
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 source content to generate analysis, storyboards, and image prompts.
- Ingestion points: Source content provided as files or text in Step 1.2.
- Capability inventory: File system access, PDF generation, and shell command execution.
- Boundary markers: Lacks explicit boundary markers or instructions to disregard embedded commands in source text.
- Sanitization: No explicit input sanitization process is defined.
- [COMMAND_EXECUTION]: The skill uses shell commands for setup checks and executes local scripts using npx and the bun runtime, specifically calling a sibling script at a relative path for image generation.
Audit Metadata