Integrations

Integrations in AgentScope AI enable seamless connectivity between agents and external services, APIs, and databases. By leveraging integrations, developers can enhance agent capabilities, automate workflows, and streamline data processing.

This document outlines the integration methods available in AgentScope AI, including API integrations, database connections, and third-party service connections.

---

Integration Types

1. API Integrations

Agents can communicate with external services via REST or GraphQL APIs. API integrations allow agents to fetch real-time data, interact with web services, and perform automated actions.

Example: Fetching Data from an External API

import { createTool } from "@agentscope/core";
import { z } from "zod";

const fetchData = async (url: string) => {
  const response = await fetch(url);
  return await response.json();
};

export const fetchAPI = createTool({
  id: "Fetch External API",
  inputSchema: z.object({
    url: z.string(),
  }),
  description: "Fetches data from an external API",
  execute: async ({ context: { url } }) => {
    return await fetchData(url);
  },
});

2. Database Integrations

Agents can interact with databases to read and write data. Supported databases include PostgreSQL, MongoDB, and MySQL.

Example: Connecting to PostgreSQL

import { Pool } from "pg";

const pool = new Pool({
  user: "user",
  host: "localhost",
  database: "agentscope",
  password: "password",
  port: 5432,
});

export const queryDatabase = async (query: string) => {
  const client = await pool.connect();
  try {
    const res = await client.query(query);
    return res.rows;
  } finally {
    client.release();
  }
};

3. Cloud & Third-Party Service Integrations

Agents can integrate with cloud services like AWS, GCP, and OpenAI, allowing them to perform computations, analyze data, and access AI-driven functionalities.

Example: Integrating with OpenAI API

import OpenAI from "openai";

const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });

export const generateText = async (prompt: string) => {
  const response = await openai.completions.create({
    model: "gpt-4",
    prompt,
    max_tokens: 100,
  });
  return response.choices[0].text;
};

---

Setting Up Integrations

1. Configuring API Keys and Credentials

Ensure that all sensitive data, such as API keys, is stored securely in environment variables.

export OPENAI_API_KEY="your-api-key-here"
export DATABASE_URL="postgres://user:password@localhost:5432/agentscope"

2. Managing Integration Settings

Integration settings can be managed via the AgentScope AI configuration file (agentscope.config.js).

module.exports = {
  integrations: {
    openai: {
      apiKey: process.env.OPENAI_API_KEY,
    },
    database: {
      url: process.env.DATABASE_URL,
    },
  },
};

3. Debugging Integrations

To debug integrations, use logging and monitoring tools such as Winston or Datadog.

import winston from "winston";

const logger = winston.createLogger({
  level: "info",
  transports: [new winston.transports.Console()],
});

logger.info("Integration initialized successfully");