reframe-social

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill defines workflows that involve executing shell commands via the video-agent CLI tool. The instructions show how to use specific subcommands for video reframing, trimming, and concatenation.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes patterns for ingesting untrusted data by processing video files provided by the user.
  • Ingestion points: Input video files (in.mp4) referenced in the video-agent command examples (SKILL.md).
  • Boundary markers: The skill does not define specific boundary markers or instructions to ignore embedded content in the video files.
  • Capability inventory: The skill uses uv run video-agent to execute various subcommands (reframe, position-grid, trim, concat) that perform file read/write operations (SKILL.md).
  • Sanitization: There is no explicit sanitization or validation of the input video content described in the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 06:10 AM
Security Audit — agent-trust-hub — reframe-social