dreamina-cli
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation in README.md and references/integration.md provides instructions for downloading the Dreamina CLI tool directly from its official domain at jimeng.jianying.com.
- [REMOTE_CODE_EXECUTION]: The provided installation instructions utilize a common pattern of piping a remote shell script from the service provider's official domain to bash to configure the local environment.
- [COMMAND_EXECUTION]: The core functionality of the skill is implemented through Python scripts that invoke the dreamina binary using the subprocess module. These wrappers provide an additional layer of parameter validation and error handling before command execution.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied natural language prompts for media generation. This is the primary intended use case for the tool, and the wrapper scripts include parameter validation to ensure arguments fall within supported ranges and choices.
- Ingestion points: prompt argument in generation scripts (text2image.py, text2video.py, etc.).
- Boundary markers: Not implemented in the wrapper layer; instructions are passed directly to the generation service.
- Capability inventory: Invocation of the dreamina CLI tool for generation, task management, and session control.
- Sanitization: The wrappers validate non-prompt parameters (like resolution and duration) but pass the generation prompts to the CLI as provided by the user.
Audit Metadata