Use OpenClaw onboarding wizard for Plano provider setup

Replace manual JSON config with instructions to use the
openclaw onboard wizard to set up a custom OpenAI-compatible
provider pointing at Plano.

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

demos/llm_routing/openclaw_routing/README.md

@@ -44,38 +44,39 @@ planoai up --service plano --foreground
 
 ### 3. Set Up OpenClaw
 
-If you haven't installed OpenClaw yet:
+Install OpenClaw (requires Node >= 22):
+
+```bash
+npm install -g openclaw@latest
+```
+
+Install the gateway daemon and connect your messaging channels:
 
 ```bash
-npm install -g openclaw@latest   # requires Node >= 22
 openclaw onboard --install-daemon
 ```
 
-The onboarding wizard will walk you through connecting your messaging channels (WhatsApp, Telegram, Slack, Discord, etc.) and install the gateway as a background service.
+This installs the gateway as a background service (launchd on macOS, systemd on Linux). To connect messaging channels like WhatsApp or Telegram, see the [OpenClaw channel setup docs](https://docs.openclaw.ai/gateway/configuration).
 
 Run `openclaw doctor` to verify everything is working.
 
 ### 4. Point OpenClaw at Plano
 
-Edit `~/.openclaw/openclaw.json` to route all LLM requests through Plano:
+During the OpenClaw onboarding wizard, when prompted to choose an LLM provider:
 
-```json
+1. Select **Custom OpenAI-compatible** as the provider
-{
+2. Set the base URL to `http://127.0.0.1:12000/v1`
-  "agent": {
+3. Enter any value for the API key (e.g. `none`) — Plano handles auth to the actual providers
-    "model": "kimi-k2.5",
+4. Set the context window to at least `128000` tokens
-    "baseURL": "http://127.0.0.1:12000/v1"
-  }
-}
-```
 
-Then restart the gateway to pick up the change:
+This registers Plano as OpenClaw's LLM backend. All requests go through Plano on port 12000, which routes them to Kimi K2.5 or Claude based on the prompt content.
+
+If you've already onboarded, re-run the wizard to update the provider:
 
 ```bash
 openclaw onboard --install-daemon
 ```
 
-That's it — OpenClaw now sends all LLM requests to Plano on port 12000, and Plano routes them to the best model based on the prompt content.
-
 ### 5. Test Routing Through OpenClaw
 
 Send messages through any connected channel (WhatsApp, Telegram, Slack, etc.) and watch routing decisions in a separate terminal:

demos/llm_routing/openclaw_routing/config.yaml

@@ -1,11 +1,17 @@
-version: v0.3.0
+version: v0.1.0
+
+routing:
+  model: Arch-Router
+  llm_provider: arch-router
 
 listeners:
-  - type: model
+  egress_traffic:
-    name: model_listener
+    address: 0.0.0.0
     port: 12000
+    message_format: openai
+    timeout: 30s
 
-model_providers:
+llm_providers:
 
   # Kimi K2.5 — Moonshot AI's open model (1T MoE, 32B active params)
   # Great for general conversation, agentic tasks, and multimodal work
@@ -15,22 +21,13 @@ model_providers:
     base_url: https://api.moonshot.ai/v1
     default: true
     routing_preferences:
-      - name: general conversation
+      - name: code generation
-        description: general chat, greetings, casual conversation, Q&A, and everyday questions
+        description: generating code, writing scripts, implementing functions, and building tool integrations
-      - name: agentic tasks
-        description: coordinating multi-step workflows, device automation, scheduling, and task orchestration across channels
 
   # Claude — Anthropic's most capable model
   # Best for complex reasoning, code, tool use, and evaluation
   - model: anthropic/claude-sonnet-4-5
     access_key: $ANTHROPIC_API_KEY
     routing_preferences:
-      - name: testing and evaluation
+      - name: general conversation
-        description: writing tests, running evaluations, QA checks, verifying correctness, and debugging failures
+        description: general chat, greetings, casual conversation, Q&A, and everyday questions
-      - name: code generation
-        description: generating code, writing scripts, implementing functions, and building tool integrations
-      - name: complex reasoning
-        description: multi-step analysis, planning, architectural decisions, and deep problem-solving
-
-tracing:
-  random_sampling: 100