The Agent Skills Directory

[SAFE]: The skill implements a secure workflow for project configuration. It performs environment checks to ensure dependencies are present before execution and uses human-in-the-loop confirmation for all critical actions.
[SAFE]: Data integrity is maintained by requiring explicit user consent before overwriting existing files and providing an automatic backup mechanism (creating .bak files) when updates occur.
[SAFE]: Although the skill reads local project files (e.g., package.json, README.md) to provide context-aware suggestions, the risk of indirect prompt injection is mitigated by the mandatory review and confirmation step in Step 3, where the user must approve the final content before it is written to the file system.

chief-init