Python SDK

daimon-sdk v0.8.0 wraps the MCP tool surface exposed by processd-mcp as grouped Python APIs such as client.files.read() and client.exec.start_session().

Requires Python 3.12+. The SDK uses httpx for HTTP transport and supports both manager-backed and direct MCP connections.

Installation

pip install daimon-sdk

PyPI: daimon-sdk

Start a Kubernetes sandbox cluster

Deploy the manager and scheduler with the Helm chart. For local validation, cd core && make e2e-k8s builds the images and deploys them into an isolated namespace in an existing kind cluster. See Sandbox cluster for the deployment model.

Quickstart

Use DaimonManagerClient when your application owns sandbox lifecycle. The context manager creates a sandbox, connects to its MCP server, and deletes it on exit by default.

import asyncio

from daimon_sdk import DaimonManagerClient


async def main() -> None:
    async with DaimonManagerClient("http://127.0.0.1:18081") as manager:
        async with manager.sandbox() as sandbox:
            runtime = await sandbox.runtime.get_context()
            print(runtime.base_workdir)

            result = await sandbox.exec.bash("python3 --version")
            print(result.display_text)

            uploaded = await sandbox.files.upload_bytes(
                "artifacts/hello.bin",
                b"hello from daimon-sdk",
            )
            print(uploaded.file_path, uploaded.bytes_written)


asyncio.run(main())

Point DaimonManagerClient at the scheduler (:18081). It is wire-compatible with the manager API — no separate scheduler client. Scheduler sandbox IDs are signed, opaque routing tokens; store and pass them back unchanged. There is no scheduler /mcp: create a sandbox first, then use mcp_url + token (or keep talking through the SDK).