The Agent Skills Directory

[SAFE]: The skill provides legitimate functionality for project management by generating and auditing vision, principles, and non-goals documentation. All operations align with the stated purpose.- [SAFE]: Tool usage is restricted to local file operations (Read, Write, Glob, Grep) and direct user interaction (AskUserQuestion). No network access, remote code downloads, or credential-harvesting patterns were found.- [SAFE]: Analysis of the instructions and reference files revealed no obfuscation, privilege escalation, or prompt injection techniques. The skill uses clearly defined modes of operation with transparent logic.- [SAFE]: The automated updating of CLAUDE.md is a standard and safe integration pattern for AI agent environments to maintain project context.

ethos