ai-image-generation
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
infshCLI tool viaBashcommands to interact with image generation models. The allowed tools are constrained to theinfshbinary. - [PROMPT_INJECTION]: The skill provides templates that interpolate user-provided text into shell commands as part of a JSON payload for remote AI models. This constitutes an indirect prompt injection surface.
- Ingestion points: User-supplied text is used in the
promptfield of the JSON input for theinfsh app runcommand in multiple examples (e.g., SKILL.md). - Boundary markers: Commands use single quotes to wrap the JSON input, providing basic shell-level separation.
- Capability inventory: The skill has the capability to execute commands using the
infshtool, which interacts with external image generation APIs. - Sanitization: There is no evidence of explicit input sanitization or validation within the skill's instructions.
- [EXTERNAL_DOWNLOADS]: The skill references external endpoints on
inference.shfor documentation and installation of the required CLI tool vianpx. These resources are consistent with the vendor's platform and the skill's stated functionality.
Audit Metadata