Gabe init

Gabe Init — KDBP project scaffolder

Gabe execution contract (E1–E7)

This skill runs under the suite execution contract — E1 EVIDENCE · E2 RUN-BEFORE-✅ · E3 NO SILENT DOWNGRADE · E4 REUSE FIRST · E5 STATE SYNC · E6 MISSING ANCHOR = STOP · E7 REPORT WHERE — floors, not ceilings; a skill's own gate may be stricter, never looser. Full text: ../gabe-docs/references/execution-contract.md (if that file is missing, E6 applies — STOP).

What this does

One-command project setup wrapping three operations: (1) creates .kdbp/ — BEHAVIOR.md, VALUES.md, DECISIONS.md, RULES.md, PENDING.md, LEDGER.md (thin session index), DOCS.md, PLAN.md, STRUCTURE.md, archive/ — plus a root CLAUDE.md, all rendered from templates under ~/.claude/templates/gabe/ (E6 STOP if a referenced template is missing — never reconstruct one from memory); PLAN.json (the machine mirror) is written later by /gabe-plan, not by init. (2) wires the 6 KDBP hooks into ~/.claude/settings.json by reading each hook object verbatim from ~/.claude/templates/gabe/hooks.json; (3) configures the project by project type (code / mockup / hybrid) and maturity (mvp / enterprise / scale), scaffolding doc stubs and, for agent apps, VALUES entries + a build checklist. Supports reset (destructive, fresh .kdbp/), update (non-destructive top-up + schema migration of an existing .kdbp/), and skip.

Procedure

1. Treat any text after the invocation as $ARGUMENTS (optional project name).
2. Read references/init-spec.md IN FULL before executing — it is the binding spec. If missing, E6 applies — STOP.
3. If .kdbp/ already exists, ask reset / update / skip and follow the mode routing table (each mode runs a fixed step sequence — reset: full create → CLAUDE.md → .gitignore seed → hooks → project type → readiness report; update: scan-missing → schema migration → hooks → condensed Update Report, skipping the full create/project-type steps).
4. On create/reset: interview for project name, one-sentence domain, maturity, project type, tech stack; render .kdbp/ files and root CLAUDE.md from the templates, substituting the interview answers; never overwrite existing user content.
5. On update: diff the existing .kdbp/ against the expected file/dir set, report present/missing/unrecognized, and on confirm create ONLY the missing items — never touch BEHAVIOR.md, VALUES.md, or any file with existing content. (The ~/.claude/gabe-arch/ lazy bootstrap is retired — gabe-arch is archived; existing user state is left untouched.)
6. Seed .gitignore with .kdbp/reviews-archive/ (idempotent, grep-before-append).
7. Check the 6 KDBP hooks in ~/.claude/settings.json; install any missing ones from ~/.claude/templates/gabe/hooks.json verbatim after a Y/n prompt — never compose hook JSON from memory.
8. Create project-type-appropriate doc stubs (architecture.md, AGENTS_USE.md, SCALING.md, architecture-patterns.md, api.md, or README.md sections depending on agent-app / web-app / CLI / library), each carrying the gabe-docs standards-reference marker.
9. Print the readiness report (reset/create) or Update Report (update mode).

Output contract (summary)

Reset/create ends with a readiness report: .kdbp/ file count, CLAUDE.md status (created/merged/preserved/backed-up/skipped), hooks installed (N/6), project type, maturity, DOCS.md mapping count, doc stubs created, and a numbered next-steps list (add VALUES.md entries, run /gabe-plan, etc.). Update mode ends with a condensed report: files added, directories created, schema migrations applied, CLAUDE.md action, hooks installed count, preserved-file count. Every ✅ reflects a file/action actually created or verified this run (E2) — nothing is claimed done without evidence. The full output contract in the spec is binding.