github-social-preview

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute various commands including image generation via a local Python script (nano_banana_pro.py) and image manipulation via ImageMagick (magick).
  • [EXTERNAL_DOWNLOADS]: The documentation references installation of well-known third-party libraries google-genai and Pillow via pip. These are standard dependencies for the described functionality.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8). It provides command templates that interpolate external repository metadata (like project name and description) into shell commands.
  • Ingestion points: Prompt templates located in SKILL.md (e.g., Quick Generation and Example Workflow sections).
  • Boundary markers: Absent; the instructions do not specify delimiters or safeguards for the interpolated metadata.
  • Capability inventory: The skill has access to the Bash tool as defined in the allowed-tools frontmatter.
  • Sanitization: No sanitization or escaping of the metadata is described before it is used in shell command strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 01:45 PM
Security Audit — agent-trust-hub — github-social-preview