Setup Project

🥷 Configure the repo so skills know where to read and write.

Scaffold the per-repo configuration that the engineering skills assume:

• Working principles — a fixed first-principles reasoning prompt written verbatim into the ## Agent skills rules block, so every agent session in the repo reasons from evidence
• Project structure — single repo, monorepo (multiple packages in one repo), or multi-repo (multiple related repos)
• Issue tracker — where issues live (GitHub, GitLab, local markdown, or other)
• Triage labels — the strings used for the five canonical triage roles
• Domain docs — where CONTEXT.md, ADRs, PRDs, and the research index live; how consumer skills find them across repos
• Documentation language — the language skills should write human-facing output in (PRDs, ADRs, CONTEXT, Issues, comments), so output is readable for the team regardless of who triggers the skill

This is a prompt-driven skill, not a deterministic script. Explore, present what you found, confirm with the user, then write.

Can be re-run. If docs/agents/ already exists, read current config first, detect what changed since last setup, present only the sections that need updating, and rewrite affected files. The user can also use this to switch issue trackers, add/remove repos from the map, or reorganize domain docs.

Relationship to the Entry Protocol