llm-goose 0.1.11

LLM testing framework for validating agent behavior and tool usage

Goose LLM 🪿

Goose is a batteries‑included Python library and CLI for validating LLM agents end‑to‑end.

Design conversational test cases, run them locally or in CI, and (optionally) plug in a React dashboard – all while staying in Python.

Turn your “vibes‑based” LLM evaluations into repeatable, versioned tests instead of “it felt smart on that one prompt” QA.

Why Goose?

Think of Goose as pytest for LLM agents:

• Stop guessing – Encode expectations once, rerun them on every model/version/deploy.
• See what actually happened – Rich execution traces, validation results, and per‑step history.
• Fits your stack – Wraps your existing agents and tools; no framework rewrite required.
• Stay in Python – Pydantic models, type hints, and a straightforward API.

Install in your project 🚀

Install the core library and CLI from PyPI:

pip install llm-goose

Quick Start: Minimal Example 🏃‍♂️

Here's a complete, runnable example of testing an LLM agent with Goose. This creates a simple weather assistant agent and tests it.

1. Set up your agent

Create my_agent.py:

from typing import Any

from dotenv import load_dotenv
from langchain.agents import create_agent
from langchain_core.messages import HumanMessage
from langchain_core.tools import tool
from goose.testing.models.messages import AgentResponse

load_dotenv()

@tool
def get_weather(location: str) -> str:
    """Get the current weather for a given location."""
    return f"The weather in {location} is sunny and 75°F."

agent = create_agent(
    model="gpt-4o-mini",
    tools=[get_weather],
    system_prompt="You are a helpful weather assistant",
)

def query_weather_agent(question: str) -> AgentResponse:
    """Query the agent and return a normalized response."""
    result = agent.invoke({"messages": [HumanMessage(content=question)]})
    return AgentResponse.from_langchain(result)

2. Set up fixtures

Create tests/conftest.py:

from goose.testing import Goose, fixture

from my_agent import query_weather_agent

@fixture(name="weather_goose") # name is optional - defaults to func name
def weather_goose_fixture() -> Goose:
    """Provide a Goose instance wired up to the sample LangChain agent."""

    return Goose(
        agent_query_func=query_weather_agent,
        validator_model=ChatOpenAI(model="gpt-4o-mini")
    )

3. Write a test

Create tests/test_weather.py. Fixture will be injected into recognized test functions. Test function and file names need to start with test_ in order to be discovered.

from goose.testing import Goose

def test_weather_query(weather_goose: Goose) -> None:
    """Test that the agent can answer weather questions."""

    weather_goose.case(
        query="What's the weather like in San Francisco?",
        expectations=[
            "Agent provides weather information for San Francisco",
            "Response mentions sunny weather and 75°F",
        ],
        expected_tool_calls=[get_weather],
    )

4. Run the test

# run help to get more information
goose-run --help

goose-run run tests

That's it! Goose will run your agent, check that it called the expected tools, and validate the response against your expectations.

Goose API & GUI

Install with extras:

pip install "llm-goose[api]"

Then launch the service from your project (for example, after wiring Goose into your own system):

# run help to get more information
goose-api --help

# start the API server (FastAPI + Uvicorn)
goose-api example_tests

React dashboard setup 🖥️

The React dashboard is a separate web application that talks to the Goose jobs API over HTTP. It is built with Vite, React, and Tailwind and is designed to be run either locally during development or deployed as a static site.

Install the published CLI from npm and let it host the built dashboard for you:

npm install -g @llm-goose/dashboard-cli

# run the dashboard
goose-dashboard

# or point the dashboard at your jobs API
GOOSE_API_URL="http://localhost:8000" goose-dashboard

For richer workflows (and the React dashboard), Goose ships an example Django system and a FastAPI jobs API. These live in this repo only – they are not installed with the PyPI package – but you can run them locally for inspiration or internal tooling.

On the dashboard, the main grid view shows one card per test in your suite, with a status pill (Passed, Failed, Queued, Running, or Not Run), the most recent duration if it has been executed, and any top‑level error from the last run. Toggling the "only failures" filter collapses the grid down to just the failing tests so you can quickly see which checks are red, which have never been executed, and which ones are currently running.

When you click into a test, the detail view shows a header with the test name, module path, latest status and duration, plus the original docstring so you remember what the scenario is meant to cover. Below that, an execution history lists each run as a card: you see every expectation with a green check or red cross, along with the validator's reasoning explaining why the run was considered a success or failure. For each step, the underlying messages are rendered as human / AI / tool bubbles, including tool calls and JSON payloads; if something went wrong mid‑run, the captured error text is shown at the bottom of the card.

Writing tests ✅

At its core, Goose lets you describe what a good interaction looks like and then assert that your agent and tools actually behave that way.

Pytest-inspired syntax

Goose cases combine a natural‑language query, human‑readable expectations, and (optionally) the tools you expect the agent to call. This example is adapted from example_tests/agent_behaviour_test.py and shows an analytical workflow where the agent both retrieves data and creates records:

def test_sale_then_inventory_update(goose_fixture: Goose) -> None:
    """Complex workflow: Sell 2 Hiking Boots and report the remaining stock."""

    count_before = Transaction.objects.count()
    inventory = ProductInventory.objects.get(product__name="Hiking Boots")
    assert inventory is not None, "Expected inventory record for Hiking Boots"

    goose_fixture.case(
        query="Sell 2 pairs of Hiking Boots to John Doe and then tell me how many we have left",
        expectations=[
            "Agent created a sale transaction for 2 Hiking Boots to John Doe",
            "Agent then checked remaining inventory after the sale",
            "Response confirmed the sale was processed",
            "Response provided updated stock information",
        ],
        expected_tool_calls=[check_inventory, create_sale],
    )

    count_after = Transaction.objects.count()
    inventory_after = ProductInventory.objects.get(product__name="Hiking Boots")

    assert count_after == count_before + 1, f"Expected 1 new transaction, got {count_after - count_before}"
    assert inventory_after is not None, "Expected inventory record after sale"
    assert inventory_after.stock == inventory.stock - 2, f"Expected stock {inventory.stock - 2}, got {inventory_after.stock}"

Custom lifecycle hooks

You can use existing lifecycle hooks or implement yours to suit your needs. Hooks are invoked before a test starts and after it finishes. This lets you setup your environment and teardown it afterwards.

from goose.testing.hooks import TestLifecycleHook

class MyLifecycleHooks(TestLifecycleHook):
    """Suite and per-test lifecycle hooks invoked around Goose executions."""

    def pre_test(self, definition: TestDefinition) -> None:
        """Hook invoked before a single test executes."""
        setup()

    def post_test(self, definition: TestDefinition) -> None:
        """Hook invoked after a single test completes."""
        teardown()


# tests/conftest.py
from goose.testing import Goose, fixture
from my_agent import query

@fixture()
def goose_fixture() -> Goose:
    """Provide a Goose instance wired up to the sample LangChain agent."""

    model = ChatOpenAI(model="gpt-4o-mini")
    return Goose(
        agent_query_func=query,
        validator_model=model,
        hooks=MyLifecycleHooks()
    )

License

MIT License – see LICENSE for full text.