Introduction

This project supports multiple methods to invoke the agents built within the system. By switching to the "Call Method" section from the sidebar, you can learn how to call Agent Party via OpenAI API, MCP server, Docker, or through the web interface.

OpenAI-style API

1. The following switch parameters have been added to the OpenAI interface:
   • enable_thinking: Default is False; whether to enable thinking mode
   • enable_deep_research: Default is False; whether to enable deep research mode
   • enable_web_search: Default is False; whether to enable web search
2. The model name defaults to super-model, which invokes the agent currently configured. If you need to call another agent, please input the corresponding agent name or ID as the model name. You can view and copy the agent name or ID on the agent interface.
3. Example calls:

from openai import OpenAI
client = OpenAI(
  api_key="super-secret-key",
  base_url="<http://127.0.0.1:3456/v1>"
)
response = client.chat.completions.create(
  model="super-model",
  messages=[
      {"role": "user", "content": "What is Super Agent Party?"}
  ]
)
print(response.choices[0].message.content)

import OpenAI from 'openai';
const client = new OpenAI({
  apiKey: "super-secret-key",
  baseURL: "<http://127.0.0.1:3456/v1>"
});
async function main() {
  const completion = await client.chat.completions.create({
      model: "super-model",
      messages: [
          { role: "user", content: "What is Super Agent Party?" }
      ]
  });
  console.log(completion.choices[0].message.content);
}
main();

curl <http://127.0.0.1:3456/v1/chat/completions> \\\\
-H "Content-Type: application/json" \\\\
-H "Authorization: Bearer super-secret-key" \\\\
-d '{
  "model": "super-model",
  "messages": [
    {"role": "user", "content": "What is Super Agent Party?"}
  ]
}'

MCP-style API

1. In any client that supports MCP, select the SSE protocol and enter the corresponding URL to use this service.
2. The API port is http://127.0.0.1:3456/mcp.
3. This interface exposes two tools:
   • get_agents: Allows external agents to query the Agent Party MCP server for a list of available agents
   • chat_with_agent_party: Enables external agents to choose an agent for conversation based on the ID returned by get_agents
4. After starting the service, write the following content into your configuration file to invoke the local MCP service:

{
  "mcpServers": {
    "super-agent-party": {
      "url": "<http://127.0.0.1:3456/mcp>"
    }
  }
}

Docker

1. Please ensure Docker is installed and the service is running. You can copy and run the following code in your terminal to launch a local instance of Super Agent Party.
2. ⭐ Note! ./super-agent-data can be replaced with any local folder. After Docker starts, all data will be cached in this local folder and will not be uploaded anywhere.
3. Example Docker startup commands: