skills/laststance/skills/gif-analyzer/Gen Agent Trust Hub

gif-analyzer

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill documentation in SKILL.md describes a process where user-supplied file paths are parsed and interpolated into a shell command for the extract_gif_frames.py script. This represents an indirect prompt injection surface where a maliciously crafted path could lead to arbitrary command execution if not properly sanitized by the agent.
  • Ingestion points: File paths provided by the user as part of the /gif command (identified in SKILL.md).
  • Boundary markers: Absent. No instructions are provided to the agent to treat the input as untrusted or to use delimiters.
  • Capability inventory: The skill utilizes shell command execution via python3 and file system writes for extracted frames and metadata.
  • Sanitization: Absent. No validation or escaping instructions for the <gif_path> variable are included in the skill.
  • [EXTERNAL_DOWNLOADS]: The skill's troubleshooting section suggests installing the Pillow library via pip. Pillow is a widely recognized and well-known image processing library.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 03:29 AM
Security Audit — agent-trust-hub — gif-analyzer