Skills
Navigate
skills/its-a-skill-issue/superpowers/ai-video-generation/Gen Agent Trust Hub

ai-video-generation

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation procedures for the belt CLI tool and additional skills from the vendor's GitHub repository (github.com/inference-sh).
  • [COMMAND_EXECUTION]: The skill uses the Bash tool, specifically constrained to the belt command via the allowed-tools configuration. This restricts the agent to a predefined set of actions provided by the inference.sh platform.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a processing layer for potentially untrusted external data.
  • Ingestion points: User-provided prompts, image URLs, audio URLs, and video URLs are ingested via the --input parameter of the belt command in SKILL.md.
  • Boundary markers: Input data is structured using JSON payloads to separate instructions from data.
  • Capability inventory: The skill is capable of executing the belt command to interface with external video generation models.
  • Sanitization: The instructions do not specify explicit sanitization or validation of the URLs or prompt strings before they are passed to the CLI.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 10:09 AM
Security Audit — agent-trust-hub — ai-video-generation