The Agent Skills Directory

[SAFE]: The skill documents standard development procedures for linting, formatting, and testing. It uses conventional project tools to ensure code quality.
[COMMAND_EXECUTION]: Provides instructions for running local project tasks using moon and bun. These commands are scoped to the project environment and do not involve remote command execution, privilege escalation, or unauthorized network activity.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent on how to verify code changes, which are externally provided inputs. The actions taken (linting, typechecking, and testing) are standard safety and verification steps for processing such inputs in a development workflow.

testing-and-verification