Containerized Agent Runtime — LangChain + FastAPI.

HAN agents run in containerized environments with persistent memory, tool calling, API integration, and prompt chaining. Standardized REST/gRPC APIs ensure consistent execution across protocols.

Agent Capabilities

HAN agents excel at real-world tasks with standardized APIs, prompt chaining, autonomous planning, and memory. Deploy specialized agents for domain-specific workflows.

LangChain Integration

Built with LangChain for composable agent architectures and tool integration.

Persistent Memory

Redis and VectorDB (Chroma, Weaviate) for memory graphs and context retention.

Multi-Step Tool Execution

LLM-powered agents with tool calling, API integration, and prompt chaining.

FastAPI Agent Runtime

Containerized agents built with FastAPI, featuring REST/gRPC APIs, WebSocket support, and standardized execution formats for seamless integration.

python

from fastapi import FastAPI
from han_agent import AgentRuntime
 
app = FastAPI()
agent = AgentRuntime(
    model="llm-model",
    memory="redis://localhost",
    tools=["search", "api_call", "database"]
)
 
@app.post("/execute")
async def execute_task(task: dict):
    result = await agent.execute_task(task)
    return {"result": result, "agent_id": agent.id}