AI Init

Set up the files agents read ("instruction surface") plus the harness loop. Tables and templates live in references/init-playbook.md — read it first. Never touch execution settings (settings.json, hooks, MCP, CI).

Step 1 — Detect

Check for agent artifacts (.claude/, CLAUDE.md, .cursor/, AGENTS.md) and coexisting frameworks (.specify/, .agent-os/, .kiro/steering/) per the playbook. Zero or multiple agents → ask one question with a recommended answer (AGENTS.md is the portable default). Never guess.

Step 2 — Inventory

List what already exists: entry file, rules directory, .ai/learnings/, .ai/backlog/, harness block, interop glue. The plan covers ONLY the missing pieces — re-running on an initialized project must be a no-op.