logicore 1.0.3

A modular, multi-provider AI agent framework for building intelligent tool-using agents with Gemini, Groq, Ollama, Azure OpenAI, and Anthropic.

Logicore AI Framework

Logicore is a powerful Python framework designed for building intelligent, multi-provider AI agents. Whether you're running local open-source models via Ollama or connecting to cloud models like Gemini and OpenAI, Logicore provides a unified, transparent, and highly robust interface without vendor lock-in.

---

🌟 Key Features

• Unified Multi-Provider Architecture: Switch between LLM backends (Ollama, Gemini, OpenAI, Groq) seamlessly. Your agent logic and tool schemas remain completely unchanged.
• Native Streaming & Reasoning Extraction: Advanced streaming support that pulls hidden <think> reasoning tokens from local models (like qwen3.5:0.8b and DeepSeek series) so your UI updates in real-time before tools execute.
• First-Class Tooling: Turn any Python function into an LLM tool automatically. Logicore parses type hints and docstrings into JSON schemas, supports **kwargs for hallucination-resilience, and safely reflects execution errors back to the model.
• Built-in Cron Job Scheduler: Endow your agents with temporal awareness. Agents can natively schedule, manage, and execute automated background tasks without external infrastructure.
• Persistent Memory & RAG: Equip agents with long-term conversational memory and semantic vector search so they never lose context across sessions.
• Built-in Skills & Copilot: Pre-packaged skill sets (Web Research, Code Review, File Manipulation) and a ready-to-use CopilotAgent for instant productivity.

---

🚀 Quickstart

Get an intelligent, tool-enabled agent running locally in two minutes.

1. Install Logicore

pip install logicore

2. Run your first Agent

Make sure you have Ollama installed and a model pulled (ollama run qwen3.5:0.8b).

import asyncio
from logicore.providers.ollama_provider import OllamaProvider
from logicore.agents.agent import Agent

# 1. Define a robust custom tool
def check_weather(location: str, **kwargs) -> str:
    """Checks the current weather for a specific location."""
    if "seattle" in location.lower():
        return "72°F and sunny."
    return "65°F and cloudy."

async def main():
    # 2. Initialize provider and agent
    provider = OllamaProvider(model_name="qwen3.5:0.8b")
    
    agent = Agent(
        llm=provider,
        role="Weather Assistant",
        system_message="Use the provided tools to answer user questions accurately.",
        tools=[check_weather],
        debug=True
    )
    
    # 3. Stream the execution live
    def on_token(token):
        print(token, end="", flush=True)

    print("Agent is thinking...\n")
    response = await agent.chat(
        "What's the weather like in Seattle today?", 
        callbacks={"on_token": on_token},
        stream=True
    )
    
    print("\n\nFinal Output:", response['content'])

if __name__ == "__main__":
    asyncio.run(main())

---

📚 Documentation

Comprehensive documentation for Logicore is available via our official site. It includes deep dives into Agents, Providers, Skills, Custom Tool guidelines, and a full API Reference.

👉 Read the Official Documentation here

---

🤝 Community & Contributions

• Discord: Join our official server to connect with other developers: Logicore Discord
• Contributing: We welcome all contributions! Please see our Contributing Guidelines to get started.

---

Built with ❤️ for multi-provider agentic workflows.