video-extend

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Potential for indirect prompt injection through processing of untrusted external content.
  • Ingestion points: The skill accepts a video_url and prompt which may originate from untrusted third-party data sources in complex agent workflows.
  • Boundary markers: Input data is encapsulated within a JSON structure passed to the CLI via the --input argument, effectively delimiting user-controlled content from the command execution context.
  • Capability inventory: The skill is restricted to the Bash(runcomfy *) tool, which allows network interaction with vendor-controlled domains and local file system writes to specified output directories.
  • Sanitization: Instructions specify that the CLI avoids shell expansion for prompt content, providing a layer of protection against command injection within the tool parameters.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:10 AM
Security Audit — agent-trust-hub — video-extend