๐ซ CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org
Python agentai-societiesartificial-intelligencecommunicative-aicooperative-aideep-learninglarge-language-modelsmulti-agent-systemsnatural-language-processing
CAMEL-AI with ACI Integration Examples
This repository contains two different approaches to integrate ACI.dev (AI Compute Infrastructure) with CAMEL-AI agents:
- Direct ACI Toolkit Integration (
aci_toolkit_camel.py) - MCP Server Integration (
camel_mcp_aci.py)
๐ Features
- Two Integration Approaches: Choose between direct ACI toolkit usage or MCP server integration
- Multiple AI Services: Access to BRAVE_SEARCH, GITHUB, and ARXIV through ACI
- Gemini Model Integration: Uses Google's Gemini 2.5 Pro model for AI responses
- Error Handling: Comprehensive error handling with detailed traceback information
๐ Project Structure
.
โโโ aci_toolkit_camel.py # Direct ACI Toolkit integration
โโโ camel_mcp_aci.py # MCP server integration approach
โโโ create_config.py # MCP configuration generator
โโโ README.md # This file
โโโ env.example # Environment variables template
๐ ๏ธ Setup
Prerequisites
- Python 3.10+
- ACI.dev API access
- Google Gemini API access
Installation
Clone the repository:
git clone https://github.com/camel-ai/camel cd camel/examples/usecases/aci_mcp/Install dependencies:
uv pip install camel-ai[all]==0.2.61 rich aci-mcpSet up environment variables:
cp env.example .env # Edit .env with your API keys and configuration
๐ง Configuration
Environment Variables
Create a .env file with the following variables:
ACI_API_KEY: Your ACI API keyGEMINI_API_KEY: Your Google Gemini API keyLINKED_ACCOUNT_OWNER_ID: Your linked account owner ID (e.g., GitHub username)
You can get these from:
- ACI API Key: Configure your apps and get your key at ACI.dev Platform
- Google API Key: Get your Gemini API key from Google AI Studio
๐ฏ Usage
Approach 1: Direct ACI Toolkit Integration
This approach uses the ACI toolkit directly within the CAMEL-AI framework:
python aci_toolkit_camel.py
Features:
- Direct integration with ACI toolkit
- Simpler setup and configuration
- Immediate access to ACI.dev tools
- Rich console output with progress indicators
Approach 2: MCP Server Integration
This approach uses MCP (Model Context Protocol) servers to provide ACI.dev functionality:
python camel_mcp_aci.py
Features:
- Uses MCP protocol for tool integration
- More modular and extensible architecture
- Interactive user input for queries
- Automatic configuration generation
- Support for multiple MCP servers
๐ง Troubleshooting
Common Issues
- Missing API Keys: Ensure all required environment variables are set in your
.envfile - ACI.dev MCP Installation: Make sure
aci-mcpis properly installed for MCP approach - Network Issues: Check internet connectivity for API calls
- Permission Issues: Ensure proper permissions for file creation (config.json)
Debug Mode
Both scripts include comprehensive error handling with traceback information for debugging.