Skip to main content

AI Agent Framework for building intelligent agents with multiple LLM providers

Project description

Demiurg

A flexible AI agent framework for building intelligent agents with support for multiple LLM providers.

Features

  • 🚀 Simple, minimal API for quick agent creation
  • 🔌 Support for multiple LLM providers (OpenAI, Anthropic, Google, etc.)
  • 📬 Built-in messaging system with conversation history
  • 📁 File handling capabilities (images, audio, text)
  • 🔧 Extensible tool system with Composio integration
  • 🏗️ Production-ready with built-in queue management
  • 🐳 Designed for both cloud and local container deployment

Installation

pip install demiurg

Quick Start

from demiurg import Agent

# Create an agent with default configuration
agent = Agent()

# Or customize your agent
agent = Agent(
    name="My Custom Agent",
    model="gpt-4",
    temperature=0.7,
    provider="openai"
)

Basic Usage

Sending Messages

from demiurg import send_text, send_file

# Send a text message
await send_text(conversation_id, "Hello from my agent!")

# Send a file
await send_file(conversation_id, "/path/to/file.png", caption="Check this out!")

Getting Conversation History

from demiurg import get_conversation_history

# Get formatted conversation history
messages = await get_conversation_history(
    conversation_id,
    limit=50,
    provider="openai"  # Format for specific LLM provider
)

Custom Agent Implementation

from demiurg import Agent

class MyAgent(Agent):
    def __init__(self):
        super().__init__(
            name="Helpful Assistant",
            model="gpt-4",
            system_prompt="You are a helpful AI assistant."
        )
    
    async def process_message(self, message, provider="openai"):
        # Add custom logic here
        response = await super().process_message(message, provider)
        return response

agent = MyAgent()

Environment Variables

The framework uses environment variables for configuration:

  • DEMIURG_BACKEND_URL: Backend API URL (default: http://backend:3000)
  • DEMIURG_AGENT_TOKEN: Authentication token
  • DEMIURG_AGENT_ID: Unique agent identifier
  • OPENAI_API_KEY: OpenAI API key (for OpenAI provider)
  • COMPOSIO_API_KEY: Composio API key (for tool integration)

Provider Support

Currently supported:

  • ✅ OpenAI (GPT-3.5, GPT-4, etc.)

Coming soon:

  • 🚧 Anthropic (Claude)
  • 🚧 Google (Gemini)
  • 🚧 Cohere
  • 🚧 Local models

Development

For development access and contribution guidelines, please contact support@demiurg.ai.

License

Copyright © 2024 Demiurg AI. All rights reserved.

This is proprietary software. See LICENSE file for details.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

demiurg-0.1.15.tar.gz (27.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

demiurg-0.1.15-py3-none-any.whl (32.9 kB view details)

Uploaded Python 3

File details

Details for the file demiurg-0.1.15.tar.gz.

File metadata

  • Download URL: demiurg-0.1.15.tar.gz
  • Upload date:
  • Size: 27.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.2

File hashes

Hashes for demiurg-0.1.15.tar.gz
Algorithm Hash digest
SHA256 5e46b7be1f0ec3c5fc4024bde4093e59032aa8173e274b9faf556bd17472dd48
MD5 89c8efb541661e51c747304420948d78
BLAKE2b-256 5a3d737ce8eca4abc828a742514080c1cf5ecc9a8d635fcd914726e69c39efdf

See more details on using hashes here.

File details

Details for the file demiurg-0.1.15-py3-none-any.whl.

File metadata

  • Download URL: demiurg-0.1.15-py3-none-any.whl
  • Upload date:
  • Size: 32.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.2

File hashes

Hashes for demiurg-0.1.15-py3-none-any.whl
Algorithm Hash digest
SHA256 24c0c30a34c4716040ced54af17f3596f1318f8ace5db5ac7d0b8d5df911e582
MD5 b53efcc29ebdbbd7f5ce2c6423d6ee5f
BLAKE2b-256 5b5a2b606eceaf0c9aa5fd3c76fb167abaec1f34679b2d4e942d312490c40fc1

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page