image-model-craft
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for utilizing standard command-line utilities such as
ffmpegto generate image montages for human review andgitto manage and compare prompt versions. It also references a project-specific tool,studio-ledger, used for tracking and reporting generation costs. - [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves compiling complex image prompts from external text sources, including character canon files and frame specifications. While this creates a surface for indirect prompt injection, the skill explicitly mandates deterministic compilation, prompt diffing, and human validation as mitigation strategies.
- [DATA_EXPOSURE]: The skill documentation includes procedures for reading from and writing to local system directories (e.g.,
~/Pictures/imagegen/_inbox,~/.studio/) and interacting with a local budgeting API (/api/v1/key). These actions are transparently presented as essential components of the creative workflow and cost-tracking requirements.
Audit Metadata