denoise-deepfilter

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions dynamically interpolate user-provided file paths into shell commands without sanitization. Ingestion points: and parameters in SKILL.md. Boundary markers: Absent. Capability inventory: Execution of deepFilter and ffprobe via shell. Sanitization: Absent. This presents a vulnerability surface for command injection if malicious file names are provided.
  • [EXTERNAL_DOWNLOADS]: The skill recommends downloading pre-compiled binaries from a third-party GitHub repository or installing dependencies via the cargo package manager.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 09:14 PM
Security Audit — agent-trust-hub — denoise-deepfilter