muapi-drone-style-video

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied text to build video generation prompts, creating an entry point for potential instructions embedded in the data.
  • Ingestion points: User-provided inputs location_or_subject and style are used to construct generation prompts in Phase A.
  • Boundary markers: The skill does not use specific delimiters or instructions to the model to ignore potential commands within the user input.
  • Capability inventory: The skill's capabilities are limited to calling a remote video generation API and using the muapi CLI; it does not have the ability to write to the local file system or execute arbitrary shell commands based on the prompt's output.
  • Sanitization: There is no evidence of filtering or escaping for the user-provided text before it is interpolated into the API request.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute specific CLI tools and network commands to fulfill its primary purpose.
  • Evidence: Usage of muapi video generate, muapi predict wait, and curl to https://api.muapi.ai/api/v1/ for generating video content. These commands are limited to the vendor's service and follow expected usage patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 05:11 PM