Quickstart

Install the SDK

pip install strivio

Add to your LangChain agent

from langchain.agents import initialize_agent
from langchain_openai import ChatOpenAI
from strivio import StrivioCallbackHandler

strivio = StrivioCallbackHandler(
    api_url="https://strivio-ai-firewall-production.up.railway.app/api/v1",
    agent_key="ak-...",
    agent_id = "my-agent-v1",
)

agent = initialize_agent(
    tools, llm,
    callbacks = [strivio]  # done.
)

Every tool call and LLM decision is now risk-scored against 16 detection categories and written to a tamper-evident audit log. This is observe-only: it records without interrupting your agent. To block high-risk interactions, see Enforcement mode below.