@lumenflow/cli 3.10.1 → 3.11.1

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lumenflow/cli",
-  "version": "3.10.1",
+  "version": "3.11.1",
   "description": "Command-line interface for LumenFlow workflow framework",
   "keywords": [
     "lumenflow",
@@ -184,13 +184,13 @@
     "xstate": "^5.28.0",
     "yaml": "^2.8.2",
     "zod": "^4.3.6",
-    "@lumenflow/agent": "3.10.1",
-    "@lumenflow/control-plane-sdk": "3.10.1",
-    "@lumenflow/initiatives": "3.10.1",
-    "@lumenflow/memory": "3.10.1",
-    "@lumenflow/kernel": "3.10.1",
-    "@lumenflow/core": "3.10.1",
-    "@lumenflow/metrics": "3.10.1"
+    "@lumenflow/agent": "3.11.1",
+    "@lumenflow/core": "3.11.1",
+    "@lumenflow/kernel": "3.11.1",
+    "@lumenflow/initiatives": "3.11.1",
+    "@lumenflow/memory": "3.11.1",
+    "@lumenflow/metrics": "3.11.1",
+    "@lumenflow/control-plane-sdk": "3.11.1"
   },
   "devDependencies": {
     "@vitest/coverage-v8": "^4.0.18",

package/packs/sidekick/.turbo/turbo-build.log

@@ -1,4 +1,4 @@
 
-> @lumenflow/packs-sidekick@3.10.0 build /home/tom/source/hellmai/lumenflow-dev/packages/@lumenflow/packs/sidekick
+> @lumenflow/packs-sidekick@3.11.1 build /home/runner/work/lumenflow-dev/lumenflow-dev/packages/@lumenflow/packs/sidekick
 > tsc
 

package/packs/sidekick/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lumenflow/packs-sidekick",
-  "version": "3.10.1",
+  "version": "3.11.1",
   "description": "Sidekick personal assistant pack for LumenFlow — 16 tools for task management, typed memory, channels, routines, and audit",
   "keywords": [
     "lumenflow",

package/packs/software-delivery/.turbo/turbo-build.log

@@ -1,4 +1,4 @@
 
-> @lumenflow/packs-software-delivery@3.10.0 build /home/tom/source/hellmai/lumenflow-dev/packages/@lumenflow/packs/software-delivery
+> @lumenflow/packs-software-delivery@3.11.1 build /home/runner/work/lumenflow-dev/lumenflow-dev/packages/@lumenflow/packs/software-delivery
 > tsc
 

package/packs/software-delivery/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lumenflow/packs-software-delivery",
-  "version": "3.10.1",
+  "version": "3.11.1",
   "description": "Software delivery pack for LumenFlow — work units, gates, lanes, initiatives, and agent coordination",
   "keywords": [
     "lumenflow",

package/templates/core/.lumenflow/constraints.md.template

@@ -194,8 +194,8 @@ If you see something broken on main (failing gates, format issues, typos, lint e
 
 - `Write(workspace.yaml, ...)` -- use `pnpm config:set` instead
 - `Edit(workspace.yaml, ...)` -- use `pnpm config:set` instead
-- `Write(docs/04-operations/tasks/wu/WU-XXX.yaml, ...)` -- use `pnpm wu:edit` instead
-- `Edit(docs/04-operations/tasks/wu/WU-XXX.yaml, ...)` -- use `pnpm wu:edit` instead
+- `Write({{DOCS_TASKS_PATH}}/wu/WU-XXX.yaml, ...)` -- use `pnpm wu:edit` instead
+- `Edit({{DOCS_TASKS_PATH}}/wu/WU-XXX.yaml, ...)` -- use `pnpm wu:edit` instead
 
 **Exception:** Reading YAML files with the Read tool is acceptable for inspection purposes.
 
@@ -288,7 +288,7 @@ git worktree prune
 
 **When approaching context limits, spawn a fresh agent instead of continuing after compaction.**
 
-Context compaction causes agents to lose critical rules. See [wu-sizing-guide.md](../docs/04-operations/_frameworks/lumenflow/wu-sizing-guide.md) for mandatory thresholds and the spawn-fresh protocol.
+Context compaction causes agents to lose critical rules. See [wu-sizing-guide.md](../{{DOCS_OPERATIONS_PATH}}/_frameworks/lumenflow/wu-sizing-guide.md) for mandatory thresholds and the spawn-fresh protocol.
 
 ---
 

package/templates/core/LUMENFLOW.md.template

@@ -4,7 +4,7 @@
 
 LumenFlow is a vendor-agnostic workflow framework for AI-native software development.
 
-> **Context Safety**: When approaching context limits (80% usage, 50+ tool calls), spawn a fresh agent instead of continuing after compaction. See [wu-sizing-guide.md](docs/04-operations/_frameworks/lumenflow/wu-sizing-guide.md).
+> **Context Safety**: When approaching context limits (80% usage, 50+ tool calls), spawn a fresh agent instead of continuing after compaction. See [wu-sizing-guide.md]({{DOCS_ONBOARDING_PATH}}/wu-sizing-guide.md).
 
 ---
 
@@ -12,7 +12,7 @@ LumenFlow is a vendor-agnostic workflow framework for AI-native software develop
 
 Completion is a **two-step process**: run `wu:prep` from the worktree (runs gates, prints copy-paste instruction), then run `wu:done` from main (merge + cleanup only). Do NOT run `wu:done` from a worktree, skip `wu:prep`, or forget to run `wu:done` after `wu:prep`.
 
-For detailed troubleshooting, common mistakes, and recovery steps, see [troubleshooting-wu-done.md](docs/04-operations/_frameworks/lumenflow/agent/onboarding/troubleshooting-wu-done.md).
+For detailed troubleshooting, common mistakes, and recovery steps, see [troubleshooting-wu-done.md]({{DOCS_ONBOARDING_PATH}}/troubleshooting-wu-done.md).
 
 ---
 
@@ -126,7 +126,7 @@ When `requireRemote: true` (default):
 7. **Command Clarity**: If you are unsure about a command, run `<command> --help` first. Do not guess.
 8. **Gates Before Done**: All gates must pass before `wu:done`
 9. **Do Not Bypass Hooks**: No `--no-verify`, fix issues properly
-10. **Always wu:done**: Complete every WU by running `pnpm wu:done`
+10. **Finish the two-step flow**: Run `pnpm wu:prep`, then run `pnpm wu:done`
 
 ---
 
@@ -166,7 +166,7 @@ For cloud agents (Codex/Claude web/CI) operating on feature branches:
 - **LUMENFLOW.md** - This file, main workflow documentation
 - **.lumenflow/constraints.md** - Non-negotiable workflow constraints
 - **.lumenflow/rules/** - Workflow rules (git-safety.md, wu-workflow.md, etc.)
-- **docs/04-operations/\_frameworks/lumenflow/agent/onboarding/** - Agent onboarding documentation
+- **{{DOCS_OPERATIONS_PATH}}/\_frameworks/lumenflow/agent/onboarding/** - Agent onboarding documentation
 
 ### Public Starlight Docs (Kernel/Packs IA)
 
@@ -203,7 +203,7 @@ The `--merge` flag uses bounded markers (`LUMENFLOW:START`/`END`) to safely inse
 
 After claiming a WU, immediately `cd worktrees/<lane>-wu-xxx` and work exclusively in the worktree. Main checkout becomes read-only -- hooks will block WU commits from main. Return to main only to run `wu:done`.
 
-For the full worktree lifecycle (parallel execution, bootstrap, isolation guarantees), see [LumenFlow Agent Capsule](docs/04-operations/_frameworks/lumenflow/lumenflow-agent-capsule.md). For the mandatory pre-write check, see [.lumenflow/constraints.md](.lumenflow/constraints.md).
+For the full worktree lifecycle (parallel execution, bootstrap, isolation guarantees), see [LumenFlow Agent Capsule]({{DOCS_OPERATIONS_PATH}}/_frameworks/lumenflow/lumenflow-agent-capsule.md). For the mandatory pre-write check, see [.lumenflow/constraints.md](.lumenflow/constraints.md).
 
 ### Vendor-Agnostic Dirty-Main Guard
 
@@ -211,7 +211,7 @@ For the full worktree lifecycle (parallel execution, bootstrap, isolation guaran
 
 - In worktree mode, commands block if main checkout has non-allowlisted dirty files
 - This includes writes from MCP tools or any vendor client that bypasses hook execution
-- Allowed dirty prefixes on main: `docs/04-operations/tasks/wu/`, `.lumenflow/`, `.claude/`, `plan/`
+- Allowed dirty prefixes on main: `{{DOCS_TASKS_PATH}}/wu/`, `.lumenflow/`, `.claude/`, `plan/`
 - `branch-pr` mode is exempt (no local worktree/main split)
 
 ---
@@ -228,7 +228,7 @@ For the full worktree lifecycle (parallel execution, bootstrap, isolation guaran
 
 ## Core Commands
 
-> **Complete CLI reference (100+ commands):** See [quick-ref-commands.md](docs/04-operations/_frameworks/lumenflow/agent/onboarding/quick-ref-commands.md). Always run `<command> --help` for the authoritative option list.
+> **Complete CLI reference (100+ commands):** See [quick-ref-commands.md]({{QUICK_REF_LINK}}). Always run `<command> --help` for the authoritative option list.
 
 | Command                   | Description                                               |
 | ------------------------- | --------------------------------------------------------- |
@@ -258,7 +258,7 @@ For the full worktree lifecycle (parallel execution, bootstrap, isolation guaran
 Commands include **context-aware validation** that checks location, WU status, and git state. When validation fails, commands provide copy-paste ready fix commands. Configure in `workspace.yaml` under `software_delivery.experimental.context_validation`.
 The Starlight CLI reference page is intentionally curated to primary commands; use `pnpm lumenflow:commands` for complete discovery.
 
-For recovery commands, state management, memory coordination, and orchestration tools, see [quick-ref-commands.md](docs/04-operations/_frameworks/lumenflow/agent/onboarding/quick-ref-commands.md).
+For recovery commands, state management, memory coordination, and orchestration tools, see [quick-ref-commands.md]({{QUICK_REF_LINK}}).
 
 ---
 
@@ -282,11 +282,11 @@ See [.lumenflow/constraints.md](.lumenflow/constraints.md) for the 9 non-negotia
 
 If you're an AI agent, read the onboarding docs:
 
-1. [docs/04-operations/\_frameworks/lumenflow/agent/onboarding/troubleshooting-wu-done.md](docs/04-operations/_frameworks/lumenflow/agent/onboarding/troubleshooting-wu-done.md) - Most common mistake
-2. [docs/04-operations/\_frameworks/lumenflow/agent/onboarding/first-wu-mistakes.md](docs/04-operations/_frameworks/lumenflow/agent/onboarding/first-wu-mistakes.md) - First WU pitfalls
-3. [docs/04-operations/\_frameworks/lumenflow/agent/onboarding/agent-safety-card.md](docs/04-operations/_frameworks/lumenflow/agent/onboarding/agent-safety-card.md) - Safety guardrails
-4. [docs/04-operations/\_frameworks/lumenflow/agent/onboarding/quick-ref-commands.md](docs/04-operations/_frameworks/lumenflow/agent/onboarding/quick-ref-commands.md) - Command reference
-5. [docs/04-operations/\_frameworks/lumenflow/agent/onboarding/test-ratchet.md](docs/04-operations/_frameworks/lumenflow/agent/onboarding/test-ratchet.md) - Test baseline ratchet pattern
+1. [{{DOCS_OPERATIONS_PATH}}/\_frameworks/lumenflow/agent/onboarding/troubleshooting-wu-done.md]({{DOCS_ONBOARDING_PATH}}/troubleshooting-wu-done.md) - Most common mistake
+2. [{{DOCS_OPERATIONS_PATH}}/\_frameworks/lumenflow/agent/onboarding/first-wu-mistakes.md]({{DOCS_ONBOARDING_PATH}}/first-wu-mistakes.md) - First WU pitfalls
+3. [{{DOCS_OPERATIONS_PATH}}/\_frameworks/lumenflow/agent/onboarding/agent-safety-card.md]({{DOCS_ONBOARDING_PATH}}/agent-safety-card.md) - Safety guardrails
+4. [{{DOCS_OPERATIONS_PATH}}/\_frameworks/lumenflow/agent/onboarding/quick-ref-commands.md]({{QUICK_REF_LINK}}) - Command reference
+5. [{{DOCS_OPERATIONS_PATH}}/\_frameworks/lumenflow/agent/onboarding/test-ratchet.md]({{DOCS_ONBOARDING_PATH}}/test-ratchet.md) - Test baseline ratchet pattern
 
 ---
 
@@ -329,10 +329,10 @@ Supported clients: `claude-code`, `codex-cli`, `cursor`, `gemini-cli`, `windsurf
 
 ## References
 
-- [LumenFlow Agent Capsule](docs/04-operations/_frameworks/lumenflow/lumenflow-agent-capsule.md) -- Full framework reference (lifecycle, lanes, gates, DoD)
-- [Quick Reference: Commands](docs/04-operations/_frameworks/lumenflow/agent/onboarding/quick-ref-commands.md) -- Complete CLI reference (100+ commands)
-- [Troubleshooting wu:done](docs/04-operations/_frameworks/lumenflow/agent/onboarding/troubleshooting-wu-done.md) -- Most common completion mistakes
+- [LumenFlow Agent Capsule]({{DOCS_OPERATIONS_PATH}}/_frameworks/lumenflow/lumenflow-agent-capsule.md) -- Full framework reference (lifecycle, lanes, gates, DoD)
+- [Quick Reference: Commands]({{QUICK_REF_LINK}}) -- Complete CLI reference (100+ commands)
+- [Troubleshooting wu:done]({{DOCS_ONBOARDING_PATH}}/troubleshooting-wu-done.md) -- Most common completion mistakes
 - [.lumenflow/constraints.md](.lumenflow/constraints.md) -- Non-negotiable rules and forbidden commands
-- [WU Sizing Guide](docs/04-operations/_frameworks/lumenflow/wu-sizing-guide.md) -- Scoping work and context safety
+- [WU Sizing Guide]({{DOCS_ONBOARDING_PATH}}/wu-sizing-guide.md) -- Scoping work and context safety
 - [Skills Index](.claude/skills/INDEX.md)
 - [Agents README](.claude/agents/README.md)

package/templates/core/UPGRADING.md.template

@@ -8,13 +8,17 @@ This guide covers upgrading LumenFlow packages in your project.
 # Check for available updates
 pnpm outdated @lumenflow/*
 
-# Update all LumenFlow packages
-pnpm update @lumenflow/cli @lumenflow/core @lumenflow/memory @lumenflow/agent @lumenflow/initiatives
+# Update all LumenFlow packages atomically
+pnpm lumenflow:upgrade --latest
 
-# Sync documentation and templates
-pnpm exec lumenflow docs:sync
+# Refresh scaffolded onboarding docs and vendor assets
+pnpm docs:sync --force
 ```
 
+Package upgrades change the runtime `wu:brief` defaults immediately. `docs:sync --force` refreshes
+scaffolded onboarding docs and supported vendor assets, but project-local `.lumenflow/templates/`
+remains optional unless you want custom overrides.
+
 ## Version Compatibility
 
 LumenFlow follows semantic versioning:
@@ -44,10 +48,10 @@ Before upgrading, review the changelog for breaking changes:
 
 ```bash
 # Update to latest versions
-pnpm update @lumenflow/cli @lumenflow/core @lumenflow/memory @lumenflow/agent @lumenflow/initiatives
+pnpm lumenflow:upgrade --latest
 
 # Or update to a specific version
-pnpm add @lumenflow/cli@^1.2.0 @lumenflow/core@^1.2.0
+pnpm lumenflow:upgrade --version 1.2.0
 ```
 
 ### 4. Sync Documentation
@@ -55,15 +59,22 @@ pnpm add @lumenflow/cli@^1.2.0 @lumenflow/core@^1.2.0
 After upgrading, sync agent onboarding documentation:
 
 ```bash
-pnpm exec lumenflow docs:sync
+pnpm docs:sync --force
 ```
 
 This updates:
 
-- `.lumenflow/` - Workflow rules and constraints
-- `.claude/` - Claude Code skills and settings (if used)
-- `.cursor/` - Cursor rules (if used)
-- Agent onboarding documentation
+- Scaffolded onboarding documentation
+- Supported vendor assets such as Claude skills
+
+It does not rewrite root workflow files such as `LUMENFLOW.md` or `.lumenflow/constraints.md`.
+
+## Defaults vs Overrides
+
+- New installs already get the current `wu:brief` defaults.
+- Existing installs get the new runtime behavior as soon as packages are upgraded.
+- Project-local `.lumenflow/templates/manifest.yaml` and `spawn-prompt/*.md` files are only needed
+  when you want to override the built-in prompt composition.
 
 ### 5. Verify Installation
 
@@ -108,6 +119,14 @@ If `docs:sync` reports conflicts:
 2. Merge manually or accept incoming changes
 3. Run `pnpm gates` to verify
 
+### `wu:brief` Still Uses Older Wording
+
+Project-local templates override the built-in defaults. Review:
+
+- `.lumenflow/templates/manifest.yaml`
+- `.lumenflow/templates/spawn-prompt/*.md`
+- `.lumenflow/templates.<client>/spawn-prompt/*.md`
+
 ## Migration Guides
 
 For major version upgrades, see version-specific migration guides:

package/templates/core/ai/onboarding/agent-safety-card.md.template

@@ -27,20 +27,20 @@ Quick reference for AI agents working in LumenFlow projects.
 | `git stash` (on main)    | Hides work       |
 | `git clean -fd`          | Deletes files    |
 | Work in main after claim | Breaks isolation |
-| Skip wu:done             | Incomplete WU    |
+| Skip wu:prep or wu:done  | Incomplete WU    |
 
 ---
 
 ## Always Do
 
-| Action                     | Why              |
-| -------------------------- | ---------------- |
-| Read WU spec first         | Understand scope |
-| cd to worktree after claim | Isolation        |
-| Write tests before code    | TDD              |
-| Run gates before wu:done   | Quality          |
-| Run wu:done                | Complete WU      |
-| Stay within code_paths     | Scope discipline |
+| Action                               | Why                               |
+| ------------------------------------ | --------------------------------- |
+| Read WU spec first                   | Understand scope                  |
+| cd to worktree after claim           | Isolation                         |
+| Write tests before code              | TDD                               |
+| Run wu:prep in the claimed workspace | Quality and lifecycle correctness |
+| Run wu:done from main                | Complete WU                       |
+| Stay within code_paths               | Scope discipline                  |
 
 ---
 
@@ -96,7 +96,7 @@ If same error happens 3 times:
 
 ```bash
 # Check lane availability
-cat docs/04-operations/tasks/status.md
+cat {{DOCS_TASKS_PATH}}/status.md
 
 # Claim a WU
 pnpm wu:claim --id WU-XXX --lane <Lane>
@@ -104,13 +104,11 @@ pnpm wu:claim --id WU-XXX --lane <Lane>
 # Work in worktree
 cd worktrees/<lane>-wu-xxx
 
-# Run gates
-pnpm gates          # Code changes
-pnpm gates --docs-only  # Docs changes
+# Prep in the claimed workspace
+pnpm wu:prep --id WU-XXX
 
-# Complete WU
-cd /path/to/main
-pnpm wu:done --id WU-XXX
+# Complete from main
+cd /path/to/main && pnpm wu:done --id WU-XXX
 ```
 
 ---
@@ -118,6 +116,7 @@ pnpm wu:done --id WU-XXX
 ## Completion Checklist
 
 - [ ] Gates pass
+- [ ] Run wu:prep in the claimed workspace
 - [ ] cd to main
 - [ ] Run wu:done
 - [ ] Verify success output

package/templates/core/ai/onboarding/first-15-mins.md.template

@@ -0,0 +1,80 @@
+# First 15 Minutes with LumenFlow
+
+**Last updated:** {{DATE}}
+
+A fast start for agents entering an existing LumenFlow project.
+
+---
+
+## Minute 0-2: Verify the Workspace
+
+```bash
+ls LUMENFLOW.md AGENTS.md workspace.yaml
+pnpm exec lumenflow doctor
+```
+
+If the doctor reports a repo-level problem, stop there and fix it before claiming work.
+
+---
+
+## Minute 2-5: Read the Required Docs
+
+1. Open `LUMENFLOW.md` for the lifecycle overview.
+2. Scan `AGENTS.md` for repo-specific workflow rules.
+3. Review `.lumenflow/constraints.md` for the non-negotiables.
+4. Open `starting-prompt.md` if you need the full onboarding flow.
+
+---
+
+## Minute 5-8: Find the Work
+
+```bash
+cat {{DOCS_TASKS_PATH}}/status.md
+ls {{DOCS_TASKS_PATH}}/wu/*.yaml | head -5
+```
+
+Read the assigned WU spec before touching files.
+
+---
+
+## Minute 8-12: Claim and Move
+
+```bash
+pnpm wu:claim --id WU-XXX --lane "Framework: CLI"
+cd worktrees/framework-cli-wu-xxx
+pnpm bootstrap
+```
+
+After `wu:claim`, all implementation work happens in the worktree, not in the main checkout.
+
+---
+
+## Minute 12-15: Start the Right Cycle
+
+For code WUs:
+
+```bash
+pnpm test -- --run
+```
+
+For docs-heavy WUs:
+
+```bash
+pnpm format {{DOCS_ONBOARDING_PATH}}/*.md
+```
+
+Before completion, always use the two-step flow:
+
+```bash
+pnpm wu:prep --id WU-XXX
+cd <project-root> && pnpm wu:done --id WU-XXX
+```
+
+---
+
+## Key Reminders
+
+- Stay in the claimed worktree.
+- Run `pnpm lumenflow:commands` for public CLI discovery, then `--help` before first use.
+- Use `pnpm wu:prep` before `pnpm wu:done`; do not jump straight to `wu:done`.
+- If context starts getting heavy, read `./wu-sizing-guide.md` before pushing further.

package/templates/core/ai/onboarding/first-wu-mistakes.md.template

@@ -254,7 +254,7 @@ pnpm wu:edit --id WU-1348 --notes "Research findings: ..."
 **Option 2: Direct file edit**
 
 ```bash
-# Edit docs/04-operations/tasks/wu/WU-1348.yaml directly
+# Edit {{DOCS_TASKS_PATH}}/wu/WU-1348.yaml directly
 ```
 
 ### Recognizing Implied Directives
@@ -311,7 +311,7 @@ sizing_estimate:
   exception_reason: All markdown documentation files, low complexity
 ```
 
-2. **Split the WU** if the estimate genuinely exceeds session capacity (see [wu-sizing-guide.md](../../wu-sizing-guide.md) section 3 for splitting patterns).
+2. **Split the WU** if the estimate genuinely exceeds session capacity (see [wu-sizing-guide.md](./wu-sizing-guide.md) section 3 for splitting patterns).
 
 **Strict mode:** Teams can enforce sizing compliance for delegated work with `--strict-sizing` on `wu:brief`. In strict mode, missing or non-compliant sizing metadata blocks the operation.
 
@@ -320,7 +320,7 @@ sizing_estimate:
 pnpm wu:brief --id WU-XXX --client claude-code --strict-sizing
 ```
 
-See the [WU Sizing Guide](../../wu-sizing-guide.md) section 1.4 for the full contract specification.
+See the [WU Sizing Guide](./wu-sizing-guide.md) section 1.4 for the full contract specification.
 
 ---
 

package/templates/core/ai/onboarding/initiative-orchestration.md.template

@@ -10,7 +10,7 @@ Step-by-step guide for agents orchestrating multi-WU initiatives. This document
 
 Before orchestrating an initiative, ensure:
 
-1. The initiative YAML exists at `docs/04-operations/tasks/initiatives/INIT-XXX.yaml`
+1. The initiative YAML exists at `{{DOCS_TASKS_PATH}}/initiatives/INIT-XXX.yaml`
 2. All WUs are linked to the initiative via `initiative:add-wu`
 3. Dependencies between WUs are defined (blocking relationships)
 4. Lane lifecycle is locked (`pnpm lane:status` shows `locked`)

package/templates/core/ai/onboarding/lane-inference.md.template

@@ -0,0 +1,64 @@
+# Lane Inference
+
+**Last updated:** {{DATE}}
+
+Lanes define which area of the system a WU belongs to. Use the lane tools before creating or reshaping work.
+
+---
+
+## The Two Inputs
+
+Lane fit is driven by:
+
+1. The WU `code_paths`
+2. The implementation description or acceptance criteria
+
+If those disagree with the assigned lane, fix the lane or split the WU.
+
+---
+
+## Runtime Checks
+
+```bash
+pnpm wu:infer-lane --id WU-XXX
+pnpm lane:status
+```
+
+Use `wu:infer-lane` for a WU-level recommendation and `lane:status` for current lifecycle state.
+
+---
+
+## Taxonomy Source
+
+Lane metadata is stored in `.lumenflow.lane-inference.yaml`.
+
+Regenerate it when taxonomy drifts from `workspace.yaml`:
+
+```bash
+pnpm lane:suggest --output .lumenflow.lane-inference.yaml
+pnpm lane:validate
+```
+
+---
+
+## Before Delivery WUs
+
+Lane lifecycle must be complete before the first delivery WU:
+
+```bash
+pnpm lane:setup
+pnpm lane:validate
+pnpm lane:lock
+```
+
+`locked` means delivery WUs can be created safely.
+
+---
+
+## Common Failure Mode
+
+If a WU touches files outside the assigned lane's code paths, do not just keep coding.
+
+- Propose a lane change if the WU is genuinely in the wrong lane.
+- Split the WU if the work spans multiple lanes.
+- Update the taxonomy only when the lane model itself is outdated.

package/templates/core/ai/onboarding/local-only.md.template

@@ -0,0 +1,56 @@
+# Local-Only Development
+
+**Last updated:** {{DATE}}
+
+Use this local-only mode when the repo has no `origin` remote yet or you are intentionally working offline.
+
+---
+
+## Raw YAML Configuration
+
+Set the workspace config under `software_delivery.git`:
+
+```yaml
+software_delivery:
+  git:
+    requireRemote: false
+```
+
+---
+
+## Safer CLI Alternative
+
+Use the config command when possible instead of editing YAML directly:
+
+```bash
+pnpm config:set --key git.requireRemote --value false
+```
+
+The `git.requireRemote` dotpath is rooted under `software_delivery`.
+
+---
+
+## Behavior Changes
+
+When `requireRemote: false`:
+
+- `wu:create` skips remote fetch requirements.
+- `wu:claim` can proceed without pushing a lane branch.
+- Local evaluation and air-gapped testing remain unblocked.
+
+When `requireRemote: true` (default):
+
+- `wu:create` and `wu:claim` expect `origin/main`.
+- Team-visible coordination happens through the remote.
+
+---
+
+## Transitioning Back to Remote Mode
+
+```bash
+pnpm config:set --key git.requireRemote --value true
+git remote add origin <url>
+git push -u origin main
+```
+
+If the repo already has the correct remote, you can simply remove the override and resume the standard workflow.