Skip to main content

​
Get Your API Key

  1. Sign up at webrun.ai
  2. Navigate to Settings → API Keys
  3. Click Create API Key and save it
enig_xxxxxxxxxxxxxxxxxxxxxxxxxxxx
Sessions cost money until terminated. Set terminateOnCompletion: true on your final task or explicitly terminate when done.

​
Run Your First Task

The simplest approach: one API call that creates a session, runs your task, and auto-terminates. 
curl -X POST https://connect.webrun.ai/start/run-task \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "taskDetails": "Go to google.com and search for Anthropic"
  }'
If the task completes within 50 seconds, you get the result immediately: 
{
  "success": true,
  "sessionId": "a1b2c3",
  "taskId": "x9y8z7",
  "type": "task_completed",
  "data": {
    "message": "Successfully searched for Anthropic on Google"
  },
  "usage": {
    "prompt_tokens": 12450,
    "completion_tokens": 3200,
    "total_tokens": 15650,
    "completion_time": 23.5,
    "cost": 0.0124
  }
}
If the task takes longer, you’ll get a polling URL: 
{
  "success": true,
  "sessionId": "a1b2c3",
  "taskId": "x9y8z7",
  "status": "pending",
  "pollUrl": "https://connect.webrun.ai/task/a1b2c3/x9y8z7"
}
Poll the URL every 2-3 seconds until the task completes. Most tasks finish in 10-40 seconds, so faster polling wastes API quota while slower polling adds latency.

​
Multi-Task Sessions

For multiple tasks in sequence, create a persistent session: 1. Create the session 
curl -X POST https://connect.webrun.ai/start/start-session \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "mode": "default",
    "initialTask": {
      "taskDetails": "Go to amazon.com",
      "startingPoint": "https://amazon.com"
    }
  }'
2. Send follow-up tasks 
curl -X POST https://connect.webrun.ai/start/send-message \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "sessionId": "SESSION_ID",
    "message": {
      "actionType": "newTask",
      "newState": "start",
      "taskDetails": "Search for wireless keyboards",
      "startingPoint": "https://amazon.com"
    }
  }'
3. Terminate when finished 
curl -X POST https://connect.webrun.ai/start/send-message \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "sessionId": "SESSION_ID",
    "message": {
      "actionType": "state",
      "newState": "terminate"
    }
  }'
Alternatively, set terminateOnCompletion: true on your last task to auto-terminate the session. Multi-task workflows guide

​
Polling for Results

When a task returns status: "pending", poll the URL until completion. 
async function runTask(taskDetails, apiKey) {
  // Start the task
  const response = await fetch("https://connect.webrun.ai/start/run-task", {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "Authorization": `Bearer ${apiKey}`
    },
    body: JSON.stringify({ taskDetails })
  });

  const data = await response.json();

  // Return immediately if complete
  if (data.type === "task_completed") {
    return data;
  }

  // Poll if still running
  if (data.status === "pending") {
    return await pollForResult(data.sessionId, data.taskId, apiKey);
  }

  throw new Error(data.message || "Task failed");
}

async function pollForResult(sessionId, taskId, apiKey) {
  const maxAttempts = 60; // 2 minutes
  const interval = 2000; // Poll every 2 seconds

  for (let i = 0; i < maxAttempts; i++) {
    const res = await fetch(
      `https://connect.webrun.ai/task/${sessionId}/${taskId}`,
      { headers: { "Authorization": `Bearer ${apiKey}` } }
    );

    const data = await res.json();

    if (data.type === "task_completed") return data;
    if (data.type === "guardrail_trigger") return data;
    if (!data.success && data.type === "failed") {
      throw new Error(data.error);
    }

    await new Promise(r => setTimeout(r, interval));
  }

  throw new Error("Task timeout after 2 minutes");
}

// Usage
const result = await runTask("Search Google for Anthropic", "enig_xxx");
console.log(result.data.message);
console.log(`Cost: $${result.usage.cost}`);
This handles both instant and long-running tasks automatically.

​
Real-Time Updates with WebSocket

For live agent thoughts and action notifications, use WebSocket. This requires more setup but gives you sub-second visibility into task execution. WebSocket integration guide

​
Next Steps

Multi-Task Workflows

Chain multiple tasks in a session

MCP Integration

Set up Claude Desktop integration

Video Streaming

Watch sessions execute in real-time

Cost Optimization

Reduce costs and improve efficiency

File Uploads

Upload files for the agent to use in sessions