skills/halt-catch-fire/mutiny/p-video/Gen Agent Trust Hub

p-video

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions from a GitHub repository associated with inference.sh and suggests adding the 'belt-sh/cli' skill. These downloads originate from well-known service endpoints.
  • [COMMAND_EXECUTION]: The skill utilizes the 'belt' command-line interface for running video generation apps. The 'allowed-tools' field correctly restricts execution to the 'belt' command, following the principle of least privilege.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data such as text prompts and media URLs to generate video content. This is the primary function of the skill.
  • Ingestion points: Input parameters within shell commands in 'SKILL.md'.
  • Boundary markers: Parameters are passed within a structured JSON object in the command line.
  • Capability inventory: Subprocess execution via the 'belt' command.
  • Sanitization: Relies on the external CLI tool and the service backend for sanitizing input prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 08:14 PM
Security Audit — agent-trust-hub — p-video