Agent Studio in Sparkforge: Build AI Agents on a Visual Canvas

Agent Studio is Sparkforge's visual environment for building intelligent AI agents. You design how an agent is triggered, how it processes information, how it decides, and what actions it performs, all on a flexible node-based canvas.

Note: AI Studio and Agent Studio are separate tools. AI Studio is a conversational builder for websites and pages. Agent Studio is a visual builder for AI agents that respond to events, interact with users, and automate tasks.

What Agent Studio does

Agents built here can interpret input, make contextual decisions, route conversations dynamically, and use tools like knowledge search or APIs. Each agent is a graph of connected nodes that defines how it starts, how it processes information, and how it completes its task.

Key benefits

• Visual AI builder. Design agent behavior on a drag-and-drop canvas.
• Event-based execution. Trigger agents on CRM events such as chat messages, form submissions, or lead tags.
• Flexible decision logic. Route conversations using AI intent detection or rule-based conditions.
• Integrated knowledge and tools. Connect your Knowledge Base, search the web, or call APIs.
• Multi-modal output. Generate text, images, audio, or video in a single flow.
• Structured data collection. Capture emails, phone numbers, and user selections during interactions.

The Agent Studio interface

Dashboard

Open AI Agents > Agent Studio to see a list of every agent in your account. From the dashboard you can view existing agents, check status, edit, duplicate, delete, or open them in the canvas.

Canvas

The canvas is where you build the agent. Drag nodes from the palette, connect them in execution order, and configure each node's inputs, prompts, tools, and outputs.

Core node types

• Trigger nodes. Start the agent on a CRM event (new chat message, form submission, tag added, scheduled time).
• AI nodes. Use a language model to interpret input, classify intent, or generate content.
• Decision nodes. Branch based on rules or AI intent.
• Tool nodes. Search the Knowledge Base, search the web, or call an external API.
• Action nodes. Send a message, update the contact, create a task, or call a webhook.
• Output nodes. Produce text, images, audio, or video deliverables.

Build your first agent

1. Go to AI Agents > Agent Studio and click Create Agent.
2. Name the agent and pick a starting template, or start from a blank canvas.
3. Add a Trigger node (for example, Chat Message Received).
4. Add an AI node to interpret intent, then a Decision node to branch on the result.
5. For knowledge questions, add a Knowledge Base tool node and feed its output into a response action.
6. For data capture, add structured prompts to collect email, phone, or selections.
7. Connect nodes end to end and configure each node's prompt, tool, or action settings.
8. Test in the built-in preview, then save and activate.

When to use Agent Studio vs Workflows

• Use Workflows for deterministic, rule-based automations (if-this-then-that across CRM events).
• Use Agent Studio when you need contextual decisions, conversational interaction, or multi-modal output that depends on what the user actually says.