skills/cawcut/skill/cawcut-vn/Gen Agent Trust Hub

cawcut-vn

Pass

Audited by Gen Agent Trust Hub on Oct 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing the @ubnt/cawcut CLI tool via the NPM registry if it is not present on the system. This is the primary tool required for the skill to perform its intended tasks.
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to perform local operations such as checking project status, importing media, and managing video project drafts. All commands are specific to the video editing workflow.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection through the processing of external files.
  • Ingestion points: Local media files (images, videos, audio) and user-supplied .srt subtitle files are processed and their metadata or content is used in the project generation process.
  • Boundary markers: The skill implements a robust mitigation by requiring explicit user approval of a generated edit plan before performing any automated project creation or modification commands.
  • Capability inventory: The skill uses the Bash tool to execute CLI commands that manage files and project structure.
  • Sanitization: The skill utilizes structured CLI arguments and predefined menu options rather than interpreting free-form text as instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 30, 2026, 01:04 AM