The Agent Skills Directory

[SAFE]: The skill provides a structured framework for organizing Claude Code configurations. All identified scripts and documentation serve legitimate administrative and productivity purposes.
[COMMAND_EXECUTION]: The skill includes a local script (scripts/inventory.ts) executed via the Bun runtime that audits project directories for configuration files. It also provides examples for setting up shell-based hooks and MCP servers, which are standard features of the platform.
[DATA_EXFILTRATION]: Documentation within the skill explicitly promotes security by providing templates that implement 'deny' rules for sensitive files, such as .env files, SSH keys, and AWS credentials, to prevent accidental exposure to the AI agent.

dotclaude-config