root-signals 1.3.0

The Python SDK for API of Root Signals

Control and Measurement for LLM automations

Root Signals SDK streamlines the evaluation of your LLM pipelines, to yield insights about their effectiveness. While many tools exist for building the pipelines themselves, quantifying their performance can be challenging.

Root Signals addresses this gap with carefully crafted universal evaluators based on cutting edge LLM research, as well as a framework for systematically adding your own.

Easily integrate with your CI/CD process for continuous monitoring to ensure the performance stays within acceptable limits.

Install

From pypi

The preferrable way of installing the SDK is from PyPI. With pip:

pip install root-signals

Quickstart

Please set your API key to environment variable ROOTSIGNALS_API_KEY, or to local .env file.

Retrieve an API key from https://app.rootsignals.ai/settings/api-keys

For example:

export ROOTSIGNALS_API_KEY=your-Root-API-key

or, if you prefer using .env file:

echo ROOTSIGNALS_API_KEY=your-Root-API-key >> .env

Minimal skill

from root import RootSignals

# Connect to the Root Signals API
client = RootSignals()

# Create a skill
skill = client.skills.create(
    """
    Classify this text into one of the following: {{categories}}
    Text: {{text}}
    """
)

# Execute it
response = skill.run(
    {
        "text": "The expectation for rate cuts has been steadily declining.",
        "categories": "Finance, Sports, Politics",
    }
)

print(response)

# "llm_output": "Finance",
# "validation": Validation(is_valid=True, validator_results=[]),
# "model": "gpt-4o",
# "execution_log_id": "9b3c713d-7bdc-4f7d-a85c-ed7d92ff4a56",
# "rendered_prompt": "Classify this text into ...",
# "cost": 5.6e-05,

Documentation

For more details, please see the main SDK documentation.

Miscellaneous notes

Versioning policy

We follow semantic versioning; to point, major versions are not guaranteed to be backwards compatible, minor versions are, and patch versions only fix bugs.