media-processing

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Privilege Escalation (HIGH): The installation instructions require the use of sudo apt-get install. Requiring administrative privileges to install system-level software is a high-risk activity for an automated agent.
  • Command Execution (MEDIUM): The skill's primary functionality relies on executing shell commands via ffmpeg, magick, and rmbg. If an attacker can influence the filenames or parameters passed to these tools (e.g., via a crafted file name), it could lead to command injection.
  • Indirect Prompt Injection (LOW):
  • Ingestion points: External media files (video, audio, images) provided as input to the tools.
  • Boundary markers: Absent. The skill does not provide instructions to sanitize inputs or use delimiters when handling user-provided file paths.
  • Capability inventory: Shell execution of ffmpeg, magick, mogrify, and rmbg-cli across all processing tasks.
  • Sanitization: Absent. There is no evidence of validation for file headers, metadata, or path strings.
  • Unverifiable Dependencies (MEDIUM): The skill installs rmbg-cli via NPM. This is an external dependency from a non-trusted source (as defined in the security policy) that is executed with system permissions.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:46 PM