maestro-video

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes natural language prompts and external media URLs provided by the user to generate video content. This represents a standard input surface for AI video production tools.
  • Ingestion points: Natural language prompt and remote file_urls in SKILL.md used for video generation.
  • Boundary markers: Not explicitly defined in the documentation examples.
  • Capability inventory: Subprocess interaction via curl for API communication.
  • Sanitization: Documentation does not specify local sanitization, relying on the vendor API to handle input safety.
  • [COMMAND_EXECUTION]: The documentation provides examples of using curl to interact with the AceDataCloud API for creating and polling video production tasks. These operations are restricted to the vendor's official domain (api.acedata.cloud).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 05:03 PM
Security Audit — agent-trust-hub — maestro-video