The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown files (SKILL.md and backend-safe-domain-modeling.md) that define a 'contract' for how an AI agent should conduct interviews and model domains.
[SAFE]: It establishes clear security boundaries by explicitly prohibiting the modification of repository documentation, implementation files, or system state during the interview process.
[SAFE]: The local recovery mechanism uses system temporary directories for transient state but includes explicit instructions to exclude credentials and sensitive dialogue from these buffers.
[SAFE]: No executable code, remote downloads, obfuscation, or credential exposure patterns were found.

and-interview-contract