ascii-video

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of subprocess to call ffmpeg for critical pipeline tasks, including audio extraction, frame sampling, and final video encoding to MP4 or GIF formats. It also executes sysctl on macOS to detect system memory for hardware-aware performance optimization.
  • [EXTERNAL_DOWNLOADS]: The skill provides integration logic for the ElevenLabs API (api.elevenlabs.io), a well-known service, to generate text-to-speech audio for video narration. This involves sending text data and a user-provided API key to the service's official endpoint.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external media, including video files, audio files, and SRT subtitles. This creates a potential attack surface where maliciously crafted files or metadata could attempt to influence the agent's behavior or command construction during the automated script generation process, though standard library guardrails for media processing are expected to be in place.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 07:35 PM
Security Audit — agent-trust-hub — ascii-video