muapi-character-story-video
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
muapiCLI andcurlto interact with external service endpoints for image and video generation. These are standard operations for the skill's stated purpose and target vendor-controlled infrastructure. - [PROMPT_INJECTION]: The skill processes untrusted user input (character descriptions and story premises) and interpolates them directly into prompts for AI models and command-line arguments. This establishes a surface for indirect prompt injection or command injection if the executing agent does not properly sanitize the inputs.
- Ingestion points: User-provided
character_descriptionandstory_premiseinputs defined inSKILL.md. - Boundary markers: Absent; inputs are directly embedded into command strings and model prompts.
- Capability inventory: Execution of
muapiCLI commands andcurlrequests toapi.muapi.ai. - Sanitization: No explicit sanitization or validation steps are defined in the instructions.
Audit Metadata