The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill utilizes node ace and npm commands. These are standard operations for building and managing AdonisJS applications. While the allowed-tools section grants significant shell capability (e.g., node ace:*), it is aligned with the stated primary purpose of the skill for project development.
[EXTERNAL_DOWNLOADS] (SAFE): The skill references standard package installation via npm init and node ace add. These operations target the official npm registry for well-known framework packages, posing no significant risk beyond standard dependency management.
[DATA_EXFILTRATION] (SAFE): Troubleshooting steps mention checking the .env file and database connectivity. This is standard procedure for local development and does not involve exfiltrating secrets or accessing unauthorized sensitive paths.
[PROMPT_INJECTION] (SAFE): The content consists purely of instructional workflows and code snippets. There are no attempts to override agent constraints, extract system prompts, or bypass safety guidelines.

adonisjs-cli