handbrake
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the HandBrakeCLI and ffprobe tools to perform video processing tasks.
- Evidence: Found in scripts/hb-encode.py, scripts/hb-preset.py, scripts/hb-scan.py, and scripts/hb_lib.py using subprocess.run and subprocess.Popen.
- Context: Commands are constructed as argument lists, which is a secure way to interface with external tools and prevents shell-level injection.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external video files and metadata, creating a surface for indirect prompt injection.
- Ingestion points: Filenames and video stream metadata are read from the filesystem and tool outputs in scripts/hb-scan.py and scripts/hb-encode.py.
- Boundary markers: No specific delimiters or warnings are used in the prompt instructions to handle untrusted data from filenames.
- Capability inventory: The skill has the ability to read and write files and execute command-line tools.
- Sanitization: Subprocess calls use argument lists to prevent shell-level command injection, reducing the impact of malicious inputs.
- [EXTERNAL_DOWNLOADS]: The skill mentions external sources for acquiring required binaries.
- Evidence: scripts/hb.sh points to the official HandBrake website (handbrake.fr) for manual installation of HandBrakeCLI.
- Context: This is a reference to the official project domain and is documented neutrally.
Audit Metadata