compress-images
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill involves processing local directory paths and updating references within content files, which represents an indirect prompt injection surface if the processed data contains malicious instructions. However, this is inherent to the skill's primary function of asset optimization. 1. Ingestion points: Target directory provided via $ARGUMENTS and project content files. 2. Boundary markers: None specified in the instructions. 3. Capability inventory: File and directory management tools (mkdir, mv, ls) and image conversion (cwebp). 4. Sanitization: No explicit input sanitization is defined for the target paths or content updates.
- [SAFE]: The skill uses standard, well-known utilities for image optimization. The use of the allowed-tools configuration appropriately limits the agent's capabilities to only the necessary bash commands, preventing unauthorized tool usage.
Audit Metadata