ai-product-photography

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the belt CLI tool to run models and manage authentication. These operations are conducted via the Bash tool.
  • [EXTERNAL_DOWNLOADS]: The instructions involve downloading the belt-sh/cli package and accessing configuration documentation from GitHub (inference-sh/skills).
  • [PROMPT_INJECTION]: The 'Batch Generation' section provides a script template that is vulnerable to indirect injection.
  • Ingestion points: User-provided strings for the PRODUCT and ANGLES variables.
  • Boundary markers: None identified in the provided script.
  • Capability inventory: Execution of subprocesses using the belt CLI.
  • Sanitization: Variable values are interpolated directly into a JSON string and passed to a shell command without escaping or validation, creating a surface for command injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 10:40 PM
Security Audit — agent-trust-hub — ai-product-photography