SprintLoop Python SDK

Official Python SDK for integrating services, data pipelines, and agents with the SprintLoop Enterprise OS.

Use this SDK to:

• Invoke agents and workflows from Python
• Build backend services on top of SprintLoop
• Send frames and events into the OS
• Collect evaluation metrics for governance

Installation

pip install sprintloop

For local development:

pip install -e .

Quickstart

import os
from sprintloop import SprintLoopClient

client = SprintLoopClient(
    api_key=os.environ.get("SPRINTLOOP_API_KEY"),
    base_url=os.environ.get("SPRINTLOOP_API_URL", "https://api.sprintloop.ai"),
)


def main() -> None:
    result = client.agents.invoke(
        agent_id="claims-intake-agent",
        input={
            "member_id": "123456789",
            "utterance": "I need to check the status of my claim.",
            "channel": "voice",
        },
    )
    print(result["output"])


if __name__ == "__main__":
    main()

Status

Early access. APIs may change before 1.0. Pin a version in production.

Related repositories

• sprintloop-enterprise-os
• sprintloop-sdk-js
• sprintloop-docs

Links

• Website: sprintloop.ai
• Docs: docs.sprintloop.ai

License

See LICENSE for details.