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), andcurl. 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 usingset -euo pipefailand 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