skills/simota/agent-skills/reel/Gen Agent Trust Hub

reel

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions provide workflows for downloading utility binaries (VHS) and font assets from trusted GitHub repositories and official package managers.- [COMMAND_EXECUTION]: The skill facilitates the execution of terminal recording tools and post-processing utilities (ffmpeg, gifsicle) to generate optimized demo artifacts.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing external data.
  • Ingestion points: Reads target CLI command output, existing tape files, and project README assets to design recording scenarios.
  • Boundary markers: None explicitly defined to delimit untrusted data from instructions.
  • Capability inventory: Shell command execution (VHS, asciinema), network operations (OpenAI/ElevenLabs TTS APIs), and file system writes.
  • Sanitization: Relies on high-level boundary instructions to prevent the inclusion of sensitive data in output recordings.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 08:29 PM
Security Audit — agent-trust-hub — reel