The Agent Skills Directory

[SAFE]: The skill performs repository discovery and template deployment using standard file operations (Read, Write, Edit, Bash, Grep, Glob). It does not execute remote code, access sensitive credentials, or perform unauthorized network operations.
[SAFE]: Instructions include explicit requirements for human-in-the-loop confirmation before overwriting or modifying existing configuration files (CLAUDE.md, AGENTS.md, or docs/agents/ content), ensuring user control over the setup process.
[SAFE]: The skill references well-known services and tools (GitHub, GitLab, Beads) in its templates as documentation for downstream agents, which is consistent with its stated purpose as a configuration manager for a multi-agent toolkit.

setup-project-skills