generate-deliverables

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses ffmpeg, ffprobe, whisper-cli, and various shell utilities (jq, awk, sed) to process video files. These operations are local and consistent with the described purpose of media production.
  • [DATA_EXPOSURE]: The skill accesses a project index file (index.json) and specific environment variables like CLAUDE_USER_DATA or XDG_DATA_HOME to resolve paths for saving outputs. This is used for project organization and does not involve exfiltration.
  • [REMOTE_CODE_EXECUTION]: The skill references an external whisper-cli but the allowed-tools frontmatter indicates it is expected to be a local executable. It does not perform any remote downloads or execution of untrusted code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:34 PM
Security Audit — agent-trust-hub — generate-deliverables