Quick Start Guide

Get your first AI agent governed by TheWARDN in under 5 minutes.

Prerequisites

• A TheWARDN account (sign up here)
• An API key (generated in Settings after signup)
• An AI agent or application that makes API calls

Step 1: Get Your API Key

1. Log in to console.thewardn.ai
2. Navigate to Settings in the sidebar
3. Copy your API key (starts with wdn_live_)

TIP

Your API key is scoped to your tenant. All agents and actions under your account share this key.

Step 2: Register an Agent

Every AI system that interacts with TheWARDN needs to be registered as an agent.

curl -X POST https://api.thewardn.ai/agents \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "my-first-agent",
    "description": "My first governed AI agent"
  }'

Response:

{
  "id": "agt_abc123xyz",
  "name": "my-first-agent",
  "description": "My first governed AI agent",
  "status": "active",
  "total_governed": 0,
  "created_at": "2026-04-10T20:00:00Z"
}

Save the id — you'll use it in every governance request.

Step 3: Govern Your First Action

Before your agent executes any action, send it to TheWARDN for governance:

curl -X POST https://api.thewardn.ai/govern \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "agent_id": "agt_abc123xyz",
    "action_type": "data_query",
    "target_service": "customer-database",
    "environment": "production",
    "confidence": {
      "incident": 0.92,
      "fix": 0.88,
      "containment": 0.95
    },
    "reasoning": "Agent needs to query customer records for support ticket #1234"
  }'

Response:

{
  "execute": true,
  "verdict": "CLEARED",
  "tier": "A",
  "seq": 1,
  "hash": "a3f2b8c9d4e5f6...",
  "message": null
}

Step 4: Handle the Verdict

Your agent should check the verdict field and act accordingly:

import requests

WARDN_API = "https://api.thewardn.ai"
API_KEY = "wdn_live_..."
AGENT_ID = "agt_abc123xyz"

def govern_action(action_type, target, confidence, reasoning=""):
    """Send action to TheWARDN for governance before executing."""
    response = requests.post(
        f"{WARDN_API}/govern",
        headers={"Authorization": f"Bearer {API_KEY}"},
        json={
            "agent_id": AGENT_ID,
            "action_type": action_type,
            "target_service": target,
            "environment": "production",
            "confidence": confidence,
            "reasoning": reasoning,
        },
    )
    result = response.json()

    if result["verdict"] == "CLEARED":
        # Safe to execute
        return True, result

    elif result["verdict"] == "HELD":
        # Action is in escrow — wait for human approval
        print(f"Action held in escrow: {result.get('escrow_id')}")
        print(f"Timeout at: {result.get('timeout_at')}")
        return False, result

    elif result["verdict"] == "BLOCKED":
        # Action is rejected — do not execute
        print(f"Action blocked: {result.get('reason')}")
        return False, result

# Usage
can_execute, verdict = govern_action(
    action_type="deploy_code",
    target="production-server",
    confidence={"incident": 0.95, "fix": 0.90, "containment": 0.92},
    reasoning="Deploying hotfix for critical bug #5678",
)

if can_execute:
    deploy_to_production()

Step 5: View Results in the Console

1. Open console.thewardn.ai
2. The Command Center shows your governance stats
3. Click Audit Trail to see the hash-chained record of your action
4. Click Live Monitor to watch actions in real-time

What's Next?

Now that you have basic governance working, explore these features:

Next Step	Description
Create Policies	Add CHAM policies to control agent behavior
Set Up Tier Mapping	Map action types to governance tiers
Apply Compliance Packs	One-click regulatory compliance
Configure L6 Model Governance	Control which AI models your agents use
Set Up Alerts	Get notified when violations occur

Important

In production, ALWAYS check the governance verdict before executing an action. An agent that ignores HELD or BLOCKED verdicts violates the governance contract and may be automatically blocked.