book-cover-design

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions for the user to install an external toolset ('npx skills add belt-sh/cli') and references documentation at 'https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md'. These are legitimate references to external resources required for the skill's functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection by demonstrating how to construct image generation commands from user inputs.
  • Ingestion points: User-provided prompt text used as input for 'belt' commands in 'SKILL.md'.
  • Boundary markers: Absent; provided examples use JSON structures but do not include specific delimiters to prevent potential instruction injection within the prompt string.
  • Capability inventory: The skill is restricted to using the 'Bash' tool specifically for 'belt' commands.
  • Sanitization: Absent; the skill relies on the user and the external API to handle prompt content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 04:09 PM
Security Audit — agent-trust-hub — book-cover-design