Quickstart | ARGUS Docs

Prerequisites

Requirements

Python 3.9 or higher
pip (or any Python package manager)
A LangGraph pipeline (or any Python callable to test with)

Installation

bash

pip install argus-agents

That's it. No extra dependencies, no config files needed to start. ARGUS ships with sensible defaults that work out of the box.

Instrument Your Graph

Wrap your LangGraph pipeline with ARGUS in three lines:

Import and create a watcher

python

from argus import ArgusWatcher

watcher = ArgusWatcher()

Attach it to your graph

Call watch()before compiling. ARGUS hooks into the graph's execution callbacks automatically.

python

watcher.watch(graph)
app = graph.compile()

Run and finalize

Run your pipeline normally. When it's done, call finalize() to trigger detection and generate the trace.

python

result = app.invoke({"messages": [("user", "What's the weather?")]})
watcher.finalize()

Run Your Pipeline

Here's a complete example — a simple LangGraph pipeline with ARGUS instrumentation:

pythonexample.py

1from argus import ArgusWatcher
2from langgraph.graph import StateGraph
3
4# 1. Create the watcher
5watcher = ArgusWatcher()
6
7# 2. Define your graph (your existing code)
8graph = StateGraph(AgentState)
9graph.add_node("agent", call_model)
10graph.add_node("tools", tool_node)
11# ... add edges ...
12
13# 3. Instrument and run
14watcher.watch(graph)
15app = graph.compile()
16result = app.invoke(initial_state)
17
18# 4. Finalize — triggers detection
19watcher.finalize()

Important

Always call watcher.finalize()after your pipeline completes. This is what triggers the detection layers and generates the trace. If you skip it, ARGUS collects raw data but doesn't analyze it.

View Results

After finalize(), you can view results in several ways:

bash

# View the latest trace in your terminal
argus trace --last

# Launch the replay UI
argus ui

# Generate a report
argus report --format html

Video coming soon

Running argus trace and viewing results in the terminal

Quick walkthrough of the ARGUS CLI trace viewer

Next Steps

Core Concepts — understand Watchers, Detectors, Traces, and Forensics
Configuration — customize detection sensitivity, enable semantic judging, configure storage
CLI Reference — all available commands and flags