LangChain + ODEI: Persistent World Models for Long-Running Agents

#ai #langchain #agents #python

LangChain + ODEI: Persistent World Models

ConversationBufferMemory resets on restart. ODEI gives LangChain agents a persistent world model.

Quick Integration

from langchain.tools import tool
import requests

@tool
def check_action(action: str) -> str:
    r = requests.post(
        "https://api.odei.ai/api/v2/guardrail/check",
        json={"action": action, "severity": "medium"}
    ).json()
    return f"{r["verdict"]}: {r.get("reasoning","")[:200]}"

@tool
def query_memory(term: str) -> str:
    r = requests.post(
        "https://api.odei.ai/api/v2/world-model/query",
        json={"queryType": "search", "searchTerm": term}
    ).json()
    return str(r)

Session Continuity

Inject world model at session start:

def build_context():
    wm = requests.get("https://api.odei.ai/api/v2/world-model/live").json()
    active = [n["title"] for n in wm["nodes"] if n["domain"] == "TACTICS"]
    return f"Current tasks: {active}"