MuleSoft Gemini Agentic EAPI

A MuleSoft Experience API that integrates with Google Gemini AI for Native Checkout Functionality. This eapi is direct plug-and-play app to enable UCP for any of the eCommerce application, enabling agentic AI capabilities within the Anypoint Platform ecosystem.

---

Overview

This API acts as an intelligent orchestration layer between enterprise systems and Google's Gemini AI, enabling:

• Agentic workflows — AI-driven decision making within Mule flows
• Natural language processing — Process unstructured data with Gemini
• Enterprise integration — Connect Gemini AI to any backend system via MuleSoft

---

Architecture

Client Request
      ↓
MuleSoft Gemini Agentic EAPI
      ↓
Google Gemini AI API
      ↓
Response with AI-generated insights

---

Getting Started

Prerequisites

• Anypoint Studio 7.15+
• Mule Runtime 4.4.0+
• Java 17
• Google Cloud account with Gemini API access
• Google Gemini API Key

Installation

1. Clone the repository:

git clone https://github.com/viplgoswami1/mulesoft-gemini-agentic-eapi.git
cd mulesoft-gemini-agentic-eapi

2. Import into Anypoint Studio:

   • File → Import → Anypoint Studio Project from File System
   • Select the cloned folder
   • Click Finish

3. Configure your credentials in src/main/resources/config.yaml:

gemini:
  api:
    key: "YOUR_GEMINI_API_KEY"
    url: "https://generativelanguage.googleapis.com/v1beta"
    model: "gemini-pro"

---

Configuration

Get a Gemini API Key

1. Go to https://ai.google.dev
2. Click Get API key in Google AI Studio
3. Create a new API key
4. Copy the key into your config

Environment Variables

Variable	Description	Required
GEMINI_API_KEY	Google Gemini API key	✅
GEMINI_MODEL	Model name (default: gemini-pro)	❌
GEMINI_MAX_TOKENS	Max response tokens (default: 1024)	❌

---

API Endpoints

POST /gemini/generate

Generate AI content using Gemini.

Request:

{
  "prompt": "Summarize the following product description for a Google Shopping listing",
  "context": "Your product description here...",
  "maxTokens": 500
}

Response:

{
  "generatedText": "AI-generated content here...",
  "model": "gemini-pro",
  "tokenCount": 245,
  "status": "SUCCESS"
}

POST /gemini/agent/execute

Execute an agentic workflow with multi-step reasoning.

Request:

{
  "task": "Analyze product catalog and suggest optimizations",
  "data": { ... },
  "steps": ["analyze", "recommend", "format"]
}

---

Example Mule Flow

<flow name="geminiAgenticFlow">
    <http:listener path="/gemini/generate" method="POST"/>
    
    <transform:message>
        <transform:set-payload><![CDATA[%dw 2.0
output application/json
---
{
    contents: [{
        parts: [{
            text: payload.prompt ++ "\n\n" ++ payload.context
        }]
    }]
}]]></transform:set-payload>
    </transform:message>
    
    <http:request 
        url="https://generativelanguage.googleapis.com/v1beta/models/gemini-pro:generateContent"
        method="POST">
        <http:headers>
            <http:header key="x-goog-api-key" value="${gemini.api.key}"/>
        </http:headers>
    </http:request>
    
    <logger message="#[payload]"/>
</flow>

---

Also See

• 🔌 Google Merchant Center MuleSoft Connector — Custom connector for Google Shopping API
• 📦 Maven Central — Published connector

---

Contributing

1. Fork the repository
2. Create your feature branch: git checkout -b feature/my-feature
3. Commit your changes: git commit -m 'feat: add my feature'
4. Push to the branch: git push origin feature/my-feature
5. Open a Pull Request

---

License

Apache License 2.0 — see LICENSE for details.

---

Author

Viplove Goswami

• GitHub: @viplgoswami1
• LinkedIn: Viplove Goswami
• Maven Central: io.github.viplgoswami1

---

Built with ❤️ for the MuleSoft and AI community