The Agent Skills Directory

[SAFE]: The skill operates as a purely instructional guide for performing data consistency audits. It does not contain executable scripts, external dependencies, or network-bound operations.
[SAFE]: The SKILL.md file defines a clear, restricted scope for a 'naming audit' discipline, emphasizing manual or scripted verification against a system's authoritative source of truth. It explicitly warns against common pitfalls like trusting stale documentation or assuming bindings based on name similarity.
[SAFE]: The references/diagram-convention.md file provides standardized formatting rules for Mermaid and terminal-based diagrams. It is a utility reference for visual consistency and contains no active code or injection vectors.
[SAFE]: Instructions for writing back to spreadsheets or creating backlog tickets are framed as high-level process steps to be performed via existing user-authorized tools (e.g., github-backlog), not as automated background actions that could lead to unauthorized data modification.

naming-audit