audio-to-music-video

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill's documentation provides bash examples that directly interpolate variables like $AUDIO, $COVER, $TITLE, and $ARTIST into shell commands and ffmpeg filters. Without explicit sanitization or quoting instructions, this creates a risk of command injection if the agent processes files with malicious names (e.g., $(rm -rf).mp3) or metadata.
  • [INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through untrusted data ingestion.
  • Ingestion points: Audio file metadata (Title, Artist) and file paths provided by the user or extracted from media files.
  • Boundary markers: Absent. There are no instructions to the agent to treat metadata as untrusted or to use delimiters to prevent instruction leakage.
  • Capability inventory: The skill has access to Bash(ffmpeg), Bash(jq), and Write operations.
  • Sanitization: Absent. The provided logic directly inserts metadata into the drawtext ffmpeg filter, which could be exploited to manipulate the video output or potentially break the filtergraph syntax to execute unexpected logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:34 PM
Security Audit — agent-trust-hub — audio-to-music-video