The Agent Skills Directory

[SAFE]: The skill is entirely documentation-based, providing technical standards for naming conventions and code structure.
[SAFE]: No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected.
[SAFE]: Code examples demonstrate standard library usage (e.g., Y.js, SQLite, IndexedDB) in a purely illustrative context.
[SAFE]: References to internal packages like @epicenter/cli and @epicenter/workspace are consistent with the vendor's provided context and represent legitimate organizational resources.

attach-primitive