Claw Setup

Connect self-hosted claw servers (OpenClaw, NVIDIA NemoClaw) to the ibl.ai platform. Run your own AI agent infrastructure and manage it through ibl.ai's APIs and applications.

Note: This guide runs against the hosted iblai.app environment. If you'd like a license to run the full ibl.ai platform locally or on your own infrastructure, reach out to our team at ibl.ai/contact.

---

Overview

This repository provides step-by-step guides for two things:

1. Server Setup -- Deploy an OpenClaw gateway on a VPS and expose it over HTTPS
2. Platform Integration -- Connect your claw server to ibl.ai and manage it through the platform's APIs and applications

Once connected, your claw instance is accessible from all ibl.ai applications -- Mentor AI, Skills AI, and any custom integration using the platform API. You configure mentors, push agent identities, assign skills, and chat with users -- all managed centrally through ibl.ai while the compute runs on your own infrastructure.

Features

• Self-hosted AI agents -- Run OpenClaw or NemoClaw on your own infrastructure while managing everything through ibl.ai
• Automatic TLS -- Caddy reverse proxy handles Let's Encrypt certificates with zero configuration
• Centralized mentor management -- Configure agent identities, personalities, and behavioral guidelines via the ibl.ai API
• Skill system -- Create reusable skills with scripts and resources, assign them to mentors, push to instances
• Multi-model support -- Use Anthropic, OpenRouter, or any OpenAI-compatible provider with automatic fallbacks
• Multi-agent deployments -- Run multiple agents on a single gateway (tutor, course creator, admissions, etc.)
• Secure by default -- Ed25519 device identity signing, loopback-only gateway binding, token-based auth
• Platform integration -- Connected instances are accessible from Mentor AI, Skills AI, and any custom integration via REST API
• Monitoring -- Health checks, connectivity tests, security audits, and version tracking through the API

Architecture

User (browser / app)
    │
    ▼
ibl.ai Platform (Django Channels / ASGI)
    │
    ▼
Claw Integration Layer (WebSocket + device identity signing)
    │
    ▼
Caddy (on your server, TLS via Let's Encrypt)
    │  reverse proxy to localhost:18789
    ▼
OpenClaw Gateway (systemd service, loopback only)
    │
    ▼
LLM Provider (Anthropic, OpenRouter, etc.)

Your server runs OpenClaw and Caddy. ibl.ai handles user-facing chat, mentor configuration, agent identity, skill management, and model provider orchestration. The platform connects to your server over a secure WebSocket with Ed25519 device identity signing.

Prerequisites

Requirement	Details
Server	VPS or dedicated, minimum 2 vCPU / 4 GB RAM (e.g. Hetzner CX22, ~$4/mo)
Domain	A domain or subdomain with a DNS A record pointing to your server's IP
LLM API key	Anthropic API key (or another provider like OpenRouter)
Firewall	Ports 80 and 443 open inbound
ibl.ai account	Platform org key and API credentials

Quick Start

1. Set up the server

SSH into your VPS, install OpenClaw, configure Caddy for TLS, and start the gateway as a systemd service.

Full server setup guide →

2. Connect to ibl.ai

Register your instance via the API, test connectivity, bind mentors, configure agents, and push config.

Full platform integration guide →

3. Chat

Open any ibl.ai application (Mentor AI, Skills AI, or your own integration) and send a message to a claw-backed mentor. Responses stream from your OpenClaw instance through the platform to the user.

Guides

Guide	Description
Server Setup	Install OpenClaw, configure Caddy, set up systemd, validate the deployment
Platform Integration	Register instance, configure mentors and agents, manage skills, API reference

Troubleshooting

See the troubleshooting section in the server setup guide and the connectivity checks in the platform integration guide.