story-bible
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill takes user-supplied sentences and briefs as input to generate content (story bibles and scripts). This data is then used to construct prompts for image and video generation models. An attacker could potentially embed malicious instructions in these briefs to influence the behavior of the downstream AI models or the agent's orchestration logic.
- [EXTERNAL_DOWNLOADS]: The documentation within the skill references downloading configuration files, model weights, and dependencies from external sources like HuggingFace and GitHub repositories of known projects (e.g., ComfyUI extensions). These references are typical for setting up the required environment for the tools described.
- [COMMAND_EXECUTION]: The skill instructions and reference materials describe the use of local command-line tools (
fuobe,seriesmith) and MCP servers to perform file operations, uploads, and trigger generation jobs. These are integral to the primary purpose of the skill as a creative asset pipeline.
Audit Metadata