skills/varghq/skills/gpt-image-gif/Gen Agent Trust Hub

gpt-image-gif

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill interacts with the vendor's official API (api.varg.ai) and storage services (s3.varg.ai) to generate and retrieve image files. These operations are consistent with the skill's stated purpose of providing an interface for the GPT Image 2.6 model.
  • [COMMAND_EXECUTION]: The skill utilizes local system binaries including ffmpeg, ffprobe, magick (ImageMagick), and curl. It relies on a bundled shell script, scripts/gif2web.sh, to perform post-processing tasks such as frame extraction and video encoding. The script follows security best practices by using set -euo pipefail and employing double quotes around variables to prevent word-splitting or basic command injection from file paths.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided prompts and images which are then used as inputs for the generative model. While this constitutes a data ingestion surface, the risk is mitigated by the structured nature of the API requests and the specialized purpose of the downstream image model.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 09:49 PM