Quick Start

This guide gets you from zero to a working AI agent in under 5 minutes. Make sure you have Synapse installed first.

Step 1 — Configure a LLM provider

Open the Synapse UI at http://localhost:3000 and navigate to Settings → LLM.

Using Ollama (no API key required)

1. Install Ollama if you haven't already
2. Pull a model: ollama pull mistral
3. In Synapse Settings, set Mode to local and Default model to mistral

Using Claude or GPT-4

1. Set Mode to cloud
2. Enter your API key (Anthropic, OpenAI, Gemini, Grok, or DeepSeek)
3. Set Default model — e.g. claude-3-5-sonnet-20241022 or gpt-4o

---

Step 2 — Create an agent

1. Click New Agent on the canvas page (or the + button in the sidebar)
2. Fill in:
   • Name — e.g. Research Agent
   • Type — conversational
   • Tools — Select web scraper or Browser Automation tool (to start)
   • System prompt — e.g. You are a helpful research assistant. When given a topic, search the web and provide a concise summary with key facts.
3. Click Save

---

Step 3 — Chat with your agent

1. Select your agent from the navbar
2. Type a message: What is the Model Context Protocol and how does it work?
3. Watch the agent reason through the problem, call tools if needed, and return a response

---

What's next?

Topic	Description
Agents	Learn about agent types, tools, memory, and system prompts
Orchestrations	Build multi-step deterministic workflows
Tools	Explore the 10 native tools available to agents
MCP Servers	Connect GitHub, Jira, Zapier, and more
API Reference	Full REST API documentation
Docker	Run Synapse in a container