OpenClaw Integration
Use UnoRouter as a model provider for your OpenClaw agent. Configure once and access every model through OpenClaw's gateway.
Overview
OpenClaw is an open-source agent platform for deploying self-hosted AI assistants across Telegram, Discord, Slack, and more from one configuration. Set its openai provider to UnoRouter's base URL and every channel reaches the full model catalog through a single OpenAI-compatible key.
Quick Config
Drop these values into the client. Your key auto-fills when you are signed in.
Base URL: https://api.unorouter.com/v1
API Key: YOUR_API_KEYCompatibility
Step-by-step setup
- 1Install and onboard
Install OpenClaw globally with npm, then run `openclaw onboard` to scaffold the config.
bashnpm install -g openclaw@latest openclaw onboard - 2Point the provider at UnoRouter
Edit `~/.openclaw/config.json` so the openai provider uses UnoRouter's base URL and reads the API key from the `OPENAI_API_KEY` env var. Set your default model under agents.
json// ~/.openclaw/config.json { "env": { "OPENAI_API_KEY": "YOUR_API_KEY" }, "agents": { "defaults": { "model": { "primary": "openai/YOUR_MODEL_ID" } } }, "providers": { "openai": { "baseUrl": "https://api.unorouter.com/v1", "apiKey": "env:OPENAI_API_KEY" } } } - 3Start OpenClaw
Run `openclaw start` to launch the agent against UnoRouter.
bashopenclaw start
# In %APPDATA%\openclaw\config.json set
# providers.openai.baseUrl to "https://api.unorouter.com/v1"
$env:OPENAI_API_KEY="YOUR_API_KEY"
openclaw onboardRecommended models
Free models that work well here. Paste any model id into the client.
Gotchas
- If OpenClaw cannot reach the API, confirm `providers.openai.baseUrl` ends in `/v1` and `OPENAI_API_KEY` is exported in the same shell.
- If the model is rejected, set `agents.defaults.model.primary` to a valid id prefixed with `openai/`, copied exactly from UnoRouter's models page.