baoyu-infographic

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from external Markdown files or user-pasted text, which is then directly interpolated into a prompt template (references/base-prompt.md) used for image generation.
  • Ingestion points: SKILL.md workflow Step 1.1 reads source content from file paths or direct user input.
  • Boundary markers: Absent; the content is placed into the {{CONTENT}} placeholder in the prompt template without delimiters or instructions to ignore embedded commands.
  • Capability inventory: The skill possesses the ability to execute shell commands via the bun runtime and perform multiple file write operations in the Assets/ directory.
  • Sanitization: No explicit sanitization or validation of the input content is mentioned.
  • [COMMAND_EXECUTION]: The skill executes a local script at .agents/skills/baoyu-image-gen/scripts/main.ts using the bun runtime. While this is the intended mechanism for image generation, it represents a privilege that could be targeted if the input content successfully subverts the generation prompt.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 03:00 PM
Security Audit — agent-trust-hub — baoyu-infographic