ai-video-generation

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTIONNO_CODE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to download an installation script from the vendor's domain at https://cli.skillsrouter.sh.
  • [REMOTE_CODE_EXECUTION]: The installation process suggests executing a remote shell script via 'curl | sh'. While this is a common distribution method for developer tools, it is a pattern of remote code execution. In this context, it originates from the skill author's infrastructure.
  • [COMMAND_EXECUTION]: The skill uses the 'sr' CLI tool via Bash to manage and run various video generation models.
  • [PROMPT_INJECTION]: The skill processes natural language prompts for video generation, which presents a surface for indirect prompt injection.
  • Ingestion points: The 'prompt' field within the JSON object passed to the 'sr' command in SKILL.md.
  • Boundary markers: None present in the provided examples.
  • Capability inventory: The skill leverages Bash to interact with the SkillsRouter CLI.
  • Sanitization: No input validation or sanitization is described in the skill metadata.
  • [NO_CODE]: The skill consists solely of a markdown configuration file (SKILL.md) and does not include any local executable scripts or binaries.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 08:44 AM