Getting Started

OpenClaw is an AI assistant that works in Telegram, Slack, WhatsApp, and other chat applications. It can browse the web, manage files, run commands, control your calendar, send emails—it actually does things across any platform you use daily.

ArmorIQ adds enterprise-grade security: every action OpenClaw takes is verified with intent tokens, cryptographic proofs, and real-time policy enforcement before execution. Whether it's sending an email, accessing files, or making API calls, ArmorIQ ensures every action is authorized and auditable.

This guide will walk you through setting up OpenClaw from source with the ArmorIQ security plugin enabled.

What You'll Build

By the end of this guide, you'll have:

• ✅ OpenClaw agent framework running locally
• ✅ ArmorIQ plugin enabled with policy enforcement
• ✅ Telegram bot integration for testing
• ✅ Complete intent verification flow working

Quick Overview

Prerequisites

Before you begin, ensure you have:

• Node.js v20 or higher
• pnpm package manager
• Git for cloning the repository
• ArmorIQ API Key from ArmorIQ Dashboard
• OpenAI or Gemini API Key for LLM access
• Telegram Account (optional but recommended for testing)

Architecture Overview

┌─────────────────────────────────────────────────────────┐
│                     Telegram Bot                        │
│                  (User Interface)                       │
└──────────────────────┬──────────────────────────────────┘
                       │
                       ▼
┌─────────────────────────────────────────────────────────┐
│                 OpenClaw Gateway                        │
│  ┌─────────────────────────────────────────────────┐   │
│  │              ArmorIQ Plugin                     │   │
│  │  • Captures LLM plan (tool steps)               │   │
│  │  • Requests intent token from IAP               │   │
│  │  • Verifies each step before execution          │   │
│  │  • Enforces policies                            │   │
│  └─────────────────────────────────────────────────┘   │
│  ┌─────────────────────────────────────────────────┐   │
│  │              Tool Execution                     │   │
│  │  • web_search, web_fetch                        │   │
│  │  • write_file, read_file                        │   │
│  │  • policy_update (if authorized)                │   │
│  └─────────────────────────────────────────────────┘   │
└──────────────────────┬──────────────────────────────────┘
                       │
                       ▼
┌─────────────────────────────────────────────────────────┐
│              ArmorIQ Backend Services                   │
│  • IAP: Issues cryptographic intent tokens & verifies   │
│  • Backend: Manages policies and audit logs             │
└─────────────────────────────────────────────────────────┘

How It Works

1. User sends prompt via Telegram to the OpenClaw bot
2. LLM creates plan with tool actions (e.g., web_search, write_file)
3. ArmorIQ captures plan and requests intent token from IAP
4. IAP issues token with cryptographic proofs for each step
5. Before each tool execution, ArmorIQ verifies:
   • Token validity
   • Step proof matches tool call
   • Active policies allow the action
6. Tool executes only if verification passes
7. Results returned to user via Telegram

Support

• Documentation: https://docs.armoriq.ai
• GitHub Issues: https://github.com/armoriq/aiq-openclaw/issues
• Email: support@armoriq.ai

Ready to get started? Proceed to Installation →