baoyu-infographic

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied content (Markdown files or pasted text) and incorporates it into a structured format for image generation. While it includes instructions to preserve data verbatim, malicious instructions embedded in the source content could potentially attempt to influence the final image generation prompt.
  • Ingestion points: Processes external Markdown files or user-pasted content as the primary input.
  • Boundary markers: Employs structured templates and headers in references/structured-content-template.md, though it does not use dedicated isolation delimiters like XML tags for the input data.
  • Capability inventory: Performs local file existence checks via Bash and invokes external image generation skills.
  • Sanitization: The skill focuses on factual extraction but does not explicitly describe a process for sanitizing input text of potential instruction-like strings before prompt construction.
  • [COMMAND_EXECUTION]: The skill uses simple Bash commands (test -f) to check for the existence of an optional configuration file (EXTEND.md) in the project directory or user home directory. This is a benign operation used for environment setup and preference loading.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 03:02 AM
Security Audit — agent-trust-hub — baoyu-infographic