mirror of
https://github.com/Sea-Haven-Industries/open-swe.git
synced 2026-09-30 06:53:14 +00:00
41 lines
1.9 KiB
Text
41 lines
1.9 KiB
Text
---
|
|
title: 'Custom Agents Overview'
|
|
description: 'Building your own agents for the Open Agent Platform'
|
|
---
|
|
|
|
# Building Your Own Agents
|
|
|
|
We built Open Agent Platform with custom agents in mind. Although we offer a few pre-built agents, we encourage you to build your own agents, and use OAP as a platform to prototype, test and use them! This guide will help you build agents that are compatible with all of Open Agent Platform's features.
|
|
|
|
## Platform Requirements
|
|
|
|
OAP is built on top of LangGraph Platform, which means all agents which you build to be used in OAP must be LangGraph agents deployed on LangGraph Platform.
|
|
|
|
## Agent Types
|
|
|
|
When building custom agents, you can create three main types:
|
|
|
|
1. **Standard Agents**: These are single-purpose agents that handle specific tasks.
|
|
2. **Tools Agents**: Agents that can access external tools via the MCP protocol.
|
|
3. **Supervisor Agents**: Agents that can orchestrate and coordinate multiple other agents.
|
|
|
|
## Development Flow
|
|
|
|
The typical development flow for creating custom agents involves:
|
|
|
|
1. Developing and testing your agent locally using LangGraph
|
|
2. Deploying your agent to LangGraph Platform
|
|
3. Configuring your Open Agent Platform to connect to your deployed agent
|
|
4. Testing and refining your agent through the OAP interface
|
|
|
|
## Getting Started
|
|
|
|
To get started with building your own agents, we recommend:
|
|
|
|
1. Familiarizing yourself with the [LangGraph framework](https://github.com/langchain-ai/langgraph)
|
|
2. Studying the example agents we provide:
|
|
- [Tools Agent](https://github.com/langchain-ai/oap-langgraph-tools-agent)
|
|
- [Supervisor Agent](https://github.com/langchain-ai/oap-agent-supervisor)
|
|
3. Understanding how to make your agent configurable, as detailed in the [Configuration](/custom-agents/configuration) section
|
|
|
|
Building custom agents allows you to create specialized AI assistants that can be easily managed and used through the Open Agent Platform interface.
|