slack-gif-creator

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • PROMPT_INJECTION (LOW): Indirect Prompt Injection surface identified through user-supplied data processing.
  • Ingestion points: The skill instructions in SKILL.md describe loading and processing user-uploaded images via PIL's Image.open() method.
  • Boundary markers: No boundary markers, delimiters, or explicit instructions to ignore embedded content are implemented in the logic.
  • Capability inventory: The skill includes file-writing capabilities via imageio.imwrite in core/gif_builder.py and performs local file reads for validation in core/validators.py.
  • Sanitization: No sanitization or content validation of the processed image data is performed to mitigate potential instruction-based attacks hidden in image pixels or metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:39 PM