baoyu-xhs-images

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes bash commands ('test -f') to detect configuration files (EXTEND.md) within the project structure and the user's home directory.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it interpolates untrusted user content into prompts for image generation tools without sufficient isolation. \n
  • Ingestion points: User-provided text content or files are ingested and processed in Step 1 of the SKILL.md workflow. \n
  • Boundary markers: The prompt assembly workflow (references/workflows/prompt-assembly.md) lacks explicit delimiters or instructions to ignore embedded commands within the interpolated content sections. \n
  • Capability inventory: The skill can write multiple files (analysis, prompts, outlines, and images) to the local file system and call external image generation skills. \n
  • Sanitization: The skill does not perform sanitization or validation of the input content before using it in the construction of prompts for downstream tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 11:41 PM