LLM Agent

A simple LLM client and AI Agent framework for OpenAI-compatible APIs.

Features

• LLM: Basic client for OpenAI-compatible APIs with streaming support
• ReActAgent: Reasoning and Acting agent (coming soon)
• PlanAndSolveAgent: Planning and problem-solving agent (coming soon)
• ReflectionAgent: Self-reflective agent (coming soon)

Installation

pip install libagent-ai

Usage

LLM Client

from libagent_ai import LLM

llm = LLM(
    model="glm-4.7",
    apiKey="your-api-key",
    baseUrl="https://api.example.com/v1"
)

messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Hello!"}
]

response = llm.think(messages)
print(response)

Using Environment Variables

Create a .env file:

LLM_API_KEY=your-api-key
LLM_BASE_URL=https://api.example.com/v1

Then use:

from libagent_ai import LLM

llm = LLM(model="glm-4.7")
response = llm.think(messages)

Development

# Install dependencies
uv sync

# Run tests
uv run python -m libagent_ai.llm

License

MIT