project-docs-management

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]:
  • The skill fetches an installation script (install.sh) and project assets from the author's official GitHub repository (github.com/devcxl/cabbage).
  • It downloads project configuration and workflow templates during initialization.
  • [REMOTE_CODE_EXECUTION]:
  • The project documentation recommends an installation flow that pipes a remote shell script from GitHub directly to the bash interpreter to set up the environment.
  • This script automates repository cloning and virtual environment configuration.
  • [COMMAND_EXECUTION]:
  • The CLI tool utilizes system commands such as git for repository management, change detection, and verifying documentation gates.
  • It uses pnpm to manage Node.js dependencies and build the integrated VitePress documentation site.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill processes Markdown files (artifacts) located in .cabbage/changes/ and docs/ (as seen in cabbage_cli/core.py).
  • Boundary markers: The system uses specific section headings (e.g., ## Goal) and task checkboxes as structural markers for validation.
  • Capability inventory: The skill can execute shell commands via subprocess to interact with Git and pnpm (as seen in cabbage_cli/cli.py and cabbage_cli/core.py).
  • Sanitization: The skill performs structural validation and placeholder detection but does not perform extensive filtering of natural language content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 06:40 PM
Security Audit — agent-trust-hub — project-docs-management