ai-marketing-videos
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
infshCLI tool to perform video generation and editing. These operations are conducted via the Bash tool to interact with remote AI models. The documentation also includes bash loops for automating the creation of multiple creative variants. - [EXTERNAL_DOWNLOADS]: The skill references the
inference.shplatform and recommends installing related agent skills from the vendor's repository. These are recognized as legitimate vendor resources and tools. - [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it processes user-provided descriptions and AI-generated scripts for downstream tasks. * Ingestion points: User-supplied prompts and intermediary script files (e.g., script.json). * Boundary markers: No delimiters or explicit instructions are provided to the agent to treat user content as data rather than instructions. * Capability inventory: Access to video and audio generation models and the ability to execute multi-step workflows. * Sanitization: Input content is passed directly to generation tools without explicit validation or escaping.
Audit Metadata