grok-video-edit

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external video files and user-supplied text prompts which are interpreted by a generative AI model (Grok) to reimagine footage.
  • Ingestion points: Video files (e.g., source.mp4, segment.mp4) and user prompts via the --prompt argument in SKILL.md.
  • Boundary markers: None specified in the instructions to delimit user content from system instructions.
  • Capability inventory: Shell command execution via ffmpeg and the video-agent CLI toolkit documented in SKILL.md.
  • Sanitization: No explicit sanitization or validation of input content is described in the workflow.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands to perform video manipulation tasks.
  • Evidence: Multiple instances of ffmpeg usage for muxing, concatenation, and filtering, as well as uv run video-agent for trimming and editing in SKILL.md.
  • [EXTERNAL_DOWNLOADS]: The skill documentation references and encourages the use of an external toolkit hosted on GitHub.
  • Evidence: Refers to https://github.com/agamm/video-agent in SKILL.md and UPSTREAM.md as the required environment for the provided scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 12:32 PM