OSS integrations

Open source agent frameworks can use boxd as their sandbox or runtime. Three integrations are live today.

Flue

Flue is the agent harness framework from Fred K. Schott, known from Astro. It feels like Claude Code, but fully headless and programmable in TypeScript. Flue agents can opt into a full sandbox, and the boxd connector gives each agent its own Linux VM:

import { Compute } from '@boxd-sh/sdk';
import { boxd } from './connectors/boxd';

const c = new Compute({ apiKey: process.env.BOXD_API_KEY });
const box = await c.box.create({ name: 'my-agent' });
const agent = await init({ sandbox: boxd(box), model: 'anthropic/claude-sonnet-4-6' });
const session = await agent.session();

You own the machine’s lifecycle through the TypeScript SDK, and the connector adapts it to Flue’s sandbox interface. The connector is written so a coding agent can install it into your Flue project, and Flue’s own docs cover it under sandboxes.

Bindu

Bindu is the identity, communication, and payments layer for AI agents. Wrap an agent written in any framework with bindufy() and it becomes a signed A2A microservice. With the boxd runtime, that microservice runs on a boxd machine instead of your laptop:

bindu deploy my_agent.py --runtime=boxd --auto-suspend=60

✓ my-agent serving at https://my-agent.boxd.sh

The agent gets its own HTTPS URL, and auto-suspend pairs naturally with it: the machine sleeps between requests and wakes when a call arrives. See the Bindu repo for the full runtime options, including deploying from a custom Docker image.

Harbor

Harbor is a framework for evaluating and optimizing AI agents and models: it runs agents like Claude Code, Codex, or OpenHands against benchmarks like Terminal-Bench and SWE-Bench across many parallel sandboxes. The boxd environment provider gives every trial its own isolated KVM microVM with a full Docker daemon, so Dockerfile and multi-container docker-compose.yaml tasks run unmodified:

export BOXD_API_KEY=<your-key>
harbor run --dataset terminal-bench@2.0 --agent claude-code --model anthropic/claude-opus-4-1 -e boxd

For benchmarks we recommend starting from snapshots: prepare a machine to your liking, snapshot it, and start every sandbox from there with --ek from_snapshot=<name>. This shortens the startup sequence drastically, because each trial restores a machine whose Docker cache already holds the task image instead of building it from scratch. The provider is awaiting merge upstream. You don’t have to wait for it: install Harbor from the pull request’s branch and everything above works today.

uv tool install "harbor[boxd] @ git+https://github.com/MichielMAnalytics/harbor.git@add-boxd-environment"

Request an integration

Building an open source framework that should run on boxd, or using one you’d like to see supported? Email contact@boxd.sh.

Getting started

Guides

Integrations & automations

Use cases

References

OSS integrations

Flue

Bindu

Harbor

Request an integration

​Flue

​Bindu

​Harbor

​Request an integration

Flue

Bindu

Harbor

Request an integration