chore: migrate CLAUDE.md to @-import rule structure (custodian integration)

Register with state-hub under custodian domain. Replace monolithic CLAUDE.md
with thin @-import index; distribute content into .claude/rules/ per ops-bridge pattern.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

.claude/rules/architecture.md

@@ -0,0 +1,26 @@
+## Architecture
+
+kaizen-agentic has two distinct layers:
+
+### 1. Python framework (`src/kaizen_agentic/`)
+
+- **`core.py`** — `Agent` (abstract base) + `AgentConfig` (dataclass). Tracks performance, supports config updates, implements kaizen interface.
+- **`optimization.py`** — `OptimizationLoop` (runs improvement cycles, detects trends, generates recommendations) + `PerformanceMetrics` (execution time, success rate, quality scores).
+
+### 2. Agent definitions (`agents/` — 17 files)
+
+Markdown instruction sets read and followed by Claude. Not executables. Naming convention: `agent-{name}.md`.
+
+| Category | Agents |
+|----------|--------|
+| Testing | `tdd-workflow`, `test-maintenance`, `testing-efficiency` |
+| Quality | `code-refactoring`, `datamodel-optimization`, `optimization` |
+| Process | `requirements-engineering`, `keepaTodofile`, `keepaChangelog`, `keepaContributingfile`, `project-management`, `priority-evaluation` |
+| Infrastructure | `setupRepository`, `tooling-optimization` |
+| Release | `releaseManager` |
+| Docs | `claude-documentation` |
+| Support | `wisdom-encouragement` |
+
+### Custodian integration
+
+The state-hub MCP resolves the agents directory via `host_paths[hostname]` → `local_path`. Tools: `list_kaizen_agents(category?)`, `get_kaizen_agent(name)`.

.claude/rules/first-session.md

@@ -0,0 +1,38 @@
+## First Session Protocol
+
+Triggered when `get_domain_summary("custodian")` shows **no workstreams**.
+The project is registered but work has not yet been structured.
+
+**Step 1 — Read, don't write**
+- `~/the-custodian/canon/projects/custodian/project_charter_v0.1.md` — purpose, scope
+- `~/the-custodian/canon/projects/custodian/roadmap_v0.1.md` — planned phases
+- Scan repo root: README, directory structure, existing code or docs
+
+**Step 2 — Survey in-progress work**
+Look for TODOs, open branches, half-finished files. Note done vs. started but incomplete.
+
+**Step 3 — Propose workstreams to Bernd**
+Propose 1–3 workstreams — each a coherent strand, weeks to months, anchored to a
+roadmap phase. **Wait for approval before creating.**
+
+**Step 4 — Create workplan file first, then DB record (ADR-001)**
+```
+workplans/kaizen-agentic-WP-NNNN-<slug>.md   ← write this first
+```
+Then register in the hub:
+```
+create_workstream(topic_id="cee7bedf-2b48-46ef-8601-006474f2ad7a", title="...", owner="...", description="...")
+create_task(workstream_id="<id>", title="...", priority="high|medium|low")
+```
+
+**Step 5 — Record the setup**
+```
+add_progress_event(
+    summary="First session: structured custodian into N workstreams, M tasks",
+    event_type="milestone",
+    topic_id="cee7bedf-2b48-46ef-8601-006474f2ad7a",
+    detail={"workstreams": [...], "tasks_created": M}
+)
+```
+
+<!-- Delete or archive this file once past first session -->

.claude/rules/repo-boundary.md

@@ -0,0 +1,8 @@
+## Repo boundary
+
+This repo owns **kaizen-agentic** only. It does not own:
+
+- State-hub MCP integration code → `the-custodian/state-hub/mcp_server/server.py`
+- Agent discovery tools (`list_kaizen_agents`, `get_kaizen_agent`) → `the-custodian`
+- Custodian coordination and workplan tracking → `the-custodian`
+- Deployment to custodiancore → `ops-bridge`

.claude/rules/repo-identity.md

@@ -0,0 +1,9 @@
+## Repo Identity
+
+**Purpose:** kaizen-agentic — AI agent development framework embracing kaizen (continuous improvement). Provides 17 specialized Claude Code companion agents plus an OptimizationLoop framework for continuous performance measurement and refinement.
+
+**Domain:** custodian
+**Repo slug:** kaizen-agentic
+**Topic ID:** cee7bedf-2b48-46ef-8601-006474f2ad7a
+
+**Custodian integration:** This repo is the single source of truth for all kaizen agents. The state-hub MCP exposes `list_kaizen_agents()` and `get_kaizen_agent(name)` tools so any connected session can discover and load agents without a local copy.

.claude/rules/session-protocol.md

@@ -0,0 +1,48 @@
+## Session Protocol
+
+State Hub: http://127.0.0.1:8000
+
+**Step 1 — Orient**
+```
+get_domain_summary("custodian")
+```
+If offline: `cd ~/the-custodian/state-hub && make api`
+
+**Step 2 — Check inbox**
+```
+get_messages(to_agent="kaizen-agentic", unread_only=True)
+```
+Mark read with `mark_message_read(message_id)`. Reply or act on coordination
+requests before proceeding.
+
+**Step 3 — Scan workplans**
+```bash
+ls workplans/
+```
+For each file with `status: active`, note pending `todo`/`in_progress` tasks.
+
+**Step 4 — Present brief**
+
+1. **Active workstreams** for `custodian` — title, task counts, blocking decisions
+2. **Pending tasks** from `workplans/` + any `[repo:kaizen-agentic]` hub tasks
+3. **Goal guidance** — if `goal_guidance` in summary:
+   - `needs_workplan`: surface as top action — *"Repo goal '{title}' has no workplan yet"*
+   - `alignment_warnings`: flag if active work is not aligned with current goal
+4. **Suggested next action** — highest-priority open item
+5. **SBOM status** — flag if `last_sbom_at` is unset for this repo
+
+If no workstreams: follow First Session Protocol (`first-session.md`).
+
+**During work:** `record_decision()` · `add_progress_event()` · `resolve_decision()`
+
+> State Hub is a *read model*. Bootstrap tools (`create_workstream`, `create_task`)
+> are First Session Protocol only. Work structure belongs in repo files (ADR-001).
+
+**Session close:**
+```
+add_progress_event(summary="...", topic_id="cee7bedf-2b48-46ef-8601-006474f2ad7a", workstream_id="<uuid>")
+```
+If workplan files were modified:
+```bash
+cd ~/the-custodian/state-hub && make fix-consistency REPO=kaizen-agentic
+```

.claude/rules/stack-and-commands.md

@@ -0,0 +1,43 @@
+## Stack and Commands
+
+**Language:** Python 3.8+
+**Package manager:** uv / pip (`.venv/`)
+**Test runner:** pytest
+**Linter/formatter:** flake8 (100-char), black (88-char), mypy (strict)
+
+### Essential commands
+
+```bash
+make setup-complete     # First-time setup: venv + package + dev deps
+source .venv/bin/activate
+make test               # Run full test suite
+make lint               # flake8 linting
+make format             # black formatting
+make clean              # Remove build artifacts
+```
+
+### TDD workflow
+
+```bash
+make tdd-start ISSUE=X  # Start issue with requirements validation
+make tdd-add-test       # Add test to current workspace
+make tdd-status         # Show workspace state
+make tdd-finish         # Move tests to main suite
+```
+
+### Issue management
+
+```bash
+make issue-list         # All issues (Gitea)
+make issue-list-open    # Open backlog
+make issue-show ISSUE=X # Issue detail
+make issue-create TITLE='...' BODY='...'
+```
+
+Run `make help` to see all available targets.
+
+### Core dependencies (pyproject.toml)
+
+- `pyyaml>=6.0` — YAML config
+- `click>=8.0.0` — CLI framework
+- `pydantic>=2.0.0` — Data validation

.claude/rules/workplan-convention.md

@@ -0,0 +1,12 @@
+## Workplan Convention (ADR-001)
+
+File location: `workplans/kaizen-agentic-WP-NNNN-<slug>.md`
+ID prefix: `KAIZEN-WP`
+
+Work items originate as files in this repo **before** being registered in the hub.
+
+Ecosystem todos from other agents arrive as `[repo:kaizen-agentic]` hub tasks —
+visible at session start. Pick one up by creating the workplan file, then registering
+the workstream.
+
+<!-- Ralph Loop rules and HEUREKA sequence: ~/.claude/CLAUDE.md — do not duplicate here -->