movie-poster-style-image-generation

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill suggests using the npx minibanana command to perform image generation tasks. The skill author includes a specific note that all such commands require explicit user approval, which aligns with security best practices for local execution.
  • [EXTERNAL_DOWNLOADS]: The skill references the minibanana@0.1.0 package, which is fetched from the npm registry at runtime. This dependency is specific to the skill's primary function of interfacing with image models via OpenRouter.
  • [PROMPT_INJECTION]: The skill processes user-defined concepts (Ingestion: [CONCEPT_DESCRIPTION] in SKILL.md). While boundary markers are absent in the prompt formula, the skill mitigates injection risks through its 'Creative Gauntlet' sanitization process, where the agent authors the final prompt components instead of passing untrusted input directly. Capability inventory includes shell execution and network access to OpenRouter.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 04:56 PM