AI Agent Automation
Getting Started

Quickstart

Build and run your first AI-powered automation workflow in minutes — fully local, privacy-first, and under your control.

This platform is designed around local AI agents and deterministic workflows. Nothing is hidden, nothing is sent to third-party SaaS platforms by default.

You define agents, compose workflows, and execute tasks using a runner that you own and operate.

1. Create an Agent

An agent represents an autonomous unit that can reason, use tools, and execute steps. Agents can send emails, read/write files, browse the web, or call language models.

{
  "name": "Research Agent",
  "description": "Summarizes and analyzes information",
  "provider": "openai",
  "model": "gpt-4o-mini",
  "capabilities": ["llm", "web_search"]
}

2. Define a Workflow

A workflow is a sequence of steps executed in order. Each step produces structured output that feeds into the next step.

{
  "name": "Daily Report Workflow",
  "metadata": {
    "steps": [
      { "stepId": "fetch", "type": "http", "method": "GET", "url": "https://api.example.com/data" },
      { "stepId": "summarize", "type": "llm", "prompt": "Summarize: {{steps.fetch.output}}" },
      { "stepId": "save", "type": "file", "action": "write", "path": "./report.txt", "content": "{{steps.summarize.output}}" },
      { "stepId": "notify", "type": "email", "to": "me@example.com", "subject": "Daily Report", "body": "{{steps.summarize.output}}" }
    ],
    "edges": [
      { "id": "e1", "source": "fetch", "target": "summarize" },
      { "id": "e2", "source": "summarize", "target": "save" },
      { "id": "e3", "source": "save", "target": "notify" }
    ]
  }
}

3. Run the Workflow

Save the workflow in the UI or via the API, then trigger a manual execution. The worker picks up the task and executes each step in order.

# Create workflow
curl -X POST http://localhost:5000/api/workflows \
  -H "Authorization: Bearer <token>" \
  -H "Content-Type: application/json" \
  -d @workflow.json

# Run workflow
curl -X POST http://localhost:5000/api/workflows/<id>/run \
  -H "Authorization: Bearer <token>"

4. Inspect Results

Check the Tasks page to see the execution status, step results, and logs. Each step's input, output, and duration are recorded for debugging.

What's Next?

  • • Explore the Agents page to configure provider-specific models.
  • • Read about Agent Memory to enable cross-execution context.
  • • Set up the Scheduler for recurring workflow runs.
  • • Configure Webhooks to trigger workflows from external services.