Quickstart · GitHub

full-tour.webm

What is Bizbox?

Enterprise-grade orchestration for AI-powered organizations

If OpenClaw is an employee, Bizbox is the company

Bizbox is an open-source, self-hosted Node.js server and React UI that orchestrates a team of AI agents to run a business — built with enterprise teams in mind. Bring your own agents, assign goals, run company-scoped agent workflows, and track your agents' work and costs from one secure, auditable dashboard.

It looks like a task manager — but under the hood it has org charts, budgets, governance, goal alignment, agent coordination, and the audit trails enterprises need.

Manage business goals, not pull requests.

	Step	Example
01	Define the goal	"Build the #1 AI note-taking app to $1M MRR."
02	Hire the team	CEO, CTO, engineers, designers, marketers — any bot, any provider.
03	Approve and run	Review strategy. Set budgets. Hit go. Monitor from the dashboard.

On the roadmap: next major areas include stronger company memory/knowledge, cloud and sandboxed agents, enterprise auth, and production-ready Kubernetes deployment. See ROADMAP.md.

Works with

OpenClaw

Claude Code

Codex

Cursor

Bash

HTTP

Google ADK

If it can receive a heartbeat, it's hired.

Bizbox is right for you if

• ✅ You want to run autonomous AI companies in an enterprise environment
• ✅ You need audit trails, governance, and cost controls that meet organizational standards
• ✅ You coordinate many different agents (OpenClaw, Codex, Claude, Cursor) toward a common goal
• ✅ You have 20 simultaneous Claude Code terminals open and lose track of what everyone is doing
• ✅ You want agents running autonomously 24/7, but still want to audit work and chime in when needed
• ✅ You want to monitor costs and enforce budgets
• ✅ You want a process for managing agents that feels like using a task manager
• ✅ You want to manage your autonomous businesses from your phone
• ✅ You need complete data isolation between teams or business units

Features

🔌 Bring Your Own Agent Any agent, any runtime, one org chart. If it can receive a heartbeat, it's hired.	🎯 Goal Alignment Every task traces back to the company mission. Agents know what to do and why.	💓 Heartbeats Agents wake on a schedule, check work, and act. Delegation flows up and down the org chart.
💰 Cost Control Monthly budgets per agent. When they hit the limit, they stop. No runaway costs.	🏢 Multi-Company One deployment, many companies. Complete data isolation. One control plane for your portfolio.	🎫 Ticket System Every conversation traced. Every decision explained. Full tool-call tracing and immutable audit log.
🛡️ Governance You're the board. Approve hires, override strategy, pause or terminate any agent — at any time.	📊 Org Chart Hierarchies, roles, reporting lines. Your agents have a boss, a title, and a job description.	📱 Mobile Ready Monitor and manage your autonomous businesses from anywhere.
🔒 Enterprise Security Self-hosted with no third-party data exposure. Full control over your infrastructure and data.	📋 Immutable Audit Trails Every agent action and board decision is logged. Meet compliance and oversight requirements.	🏗️ On-Premise Deployment Deploy on your own infrastructure — bare metal, VM, Docker, or Kubernetes. No vendor lock-in.

Problems Bizbox solves

Without Bizbox	With Bizbox
❌ You have 20 Claude Code tabs open and can't track which one does what. On reboot you lose everything.	✅ Tasks are ticket-based, conversations are threaded, sessions persist across reboots.
❌ You manually gather context from several places to remind your bot what you're actually doing.	✅ Context flows from the task up through the project and company goals — your agent always knows what to do and why.
❌ Folders of agent configs are disorganized and you're re-inventing task management, communication, and coordination between agents.	✅ Bizbox gives you org charts, ticketing, delegation, and governance out of the box — so you run a company, not a pile of scripts.
❌ Runaway loops waste hundreds of dollars of tokens and max your quota before you even know what happened.	✅ Cost tracking surfaces token budgets and throttles agents when they're out. Management prioritizes with budgets.
❌ You have recurring jobs (customer support, social, reports) and have to remember to manually kick them off.	✅ Heartbeats handle regular work on a schedule. Management supervises.
❌ You have an idea, you have to find your repo, fire up Claude Code, keep a tab open, and babysit it.	✅ Add a task in Bizbox. Your coding agent works on it until it's done. Management reviews their work.
❌ No audit trail when something goes wrong. No way to prove who authorized what.	✅ Every action is logged in an immutable activity audit trail. Every agent action and board decision is traceable.
❌ Sensitive company data leaves your infrastructure when using cloud-hosted AI orchestration tools.	✅ Fully self-hosted. Your data stays on your infrastructure. No external orchestration service has access.

Why Bizbox is special

Bizbox handles the hard orchestration details correctly — and adds the enterprise-grade properties teams need.

Atomic execution.	Task checkout and budget enforcement are atomic, so no double-work and no runaway spend.
Persistent agent state.	Agents resume the same task context across heartbeats instead of restarting from scratch.
Runtime skill injection.	Agents can learn Bizbox workflows and project context at runtime, without retraining.
Governance with rollback.	Approval gates are enforced, config changes are revisioned, and bad changes can be rolled back safely.
Goal-aware execution.	Tasks carry full goal ancestry so agents consistently see the "why," not just a title.
Portable company templates.	Export/import orgs, agents, and skills with secret scrubbing and collision handling.
True multi-company isolation.	Every entity is company-scoped, so one deployment can run many companies with separate data and audit trails.
Immutable audit trail.	Every agent action and board decision is durably logged — enabling compliance, forensics, and oversight.
Self-hosted and air-gappable.	Run on your own infrastructure with no external dependencies. Data never leaves your environment.

What Bizbox is not

Not a chatbot.	Agents have jobs, not chat windows.
Not an agent framework.	We don't tell you how to build agents. We tell you how to run a company made of them.
Not a workflow builder.	No drag-and-drop pipelines. Bizbox models companies — with org charts, goals, budgets, and governance. Company-scoped Workflows exist, but they are not a general pipeline builder.
Not a prompt manager.	Agents bring their own prompts, models, and runtimes. Bizbox manages the organization they work in.
Not a single-agent tool.	This is for teams. If you have one agent, you probably don't need Bizbox. If you have twenty — you definitely do.
Not a code review tool.	Bizbox orchestrates work, not pull requests. Bring your own review process.

Quickstart

Open source. Self-hosted. No Bizbox account required.

git clone https://github.com/zesthq/bizbox.git
cd bizbox
pnpm install
pnpm dev

This starts the API server at http://localhost:3100. An embedded PostgreSQL database is created automatically — no setup required.

Requirements: Node.js 20+, pnpm 9.15+

Authenticated / Enterprise Mode

For multi-user or network-accessible deployments:

pnpm dev --bind lan
# or for Tailscale:
pnpm dev --bind tailnet

See doc/DOCKER.md for production Docker deployments and doc/DEVELOPING.md for the full development guide.

Deploy To Fly

This fork includes a Fly configuration for the bizbox app in fly.toml. Fly needs Bizbox to run in authenticated mode because the server binds to 0.0.0.0; local_trusted is only valid for loopback/local desktop use.

If you want a private Fly deployment with no public HTTP service and plan to reach it via fly proxy, use fly.private.toml instead. It keeps the same app/runtime settings but omits [http_service].

First-time setup:

cp .env.example .env
# edit .env and set ORG=your-fly-org
make bootstrap

That creates the Fly app, provisions Fly Postgres, creates the persistent /paperclip volume, and sets the required auth secret plus public URL. The Makefile now loads .env, requires ORG for Fly bootstrap steps, passes Fly's non-interactive flags, and skips resources that already exist, so it is safe to rerun after a partial setup. If BETTER_AUTH_SECRET is already present, bootstrap preserves it instead of rotating sessions.

Defaults:

Make variable	Default
APP	bizbox
DB	bizbox-db
ORG	Required
REGION	syd
DB_VOL_GB	10
DB_MEMORY_MB	2048
VOLUME	paperclip_data
VOL_GB	10

Set ORG in .env for the easiest setup flow, or override variables inline when needed:

make bootstrap APP=my-paperclip ORG=my-fly-org

If you prefer to run the steps manually:

make fly-setup
make fly-db
make fly-volume
make fly-secrets

Deploy:

make deploy
# or deploy the private variant:
fly deploy --app bizbox --config fly.private.toml

After the first successful deploy, open the first instance admin invite:

make admin-invite

This runs pnpm paperclipai auth bootstrap-ceo inside the Fly machine via fly ssh console and prints a one-time invite URL. Open that URL in your browser to claim the first admin account. You only need this while the instance says Instance setup required.

If deploy reports that the app was not found, the Fly app has not been created in the current Fly account/org yet. Run make bootstrap once first. If the bizbox app name is unavailable or you already created a different app, update APP in Makefile and app/BIZBOX_PUBLIC_URL in fly.toml, then run make bootstrap.

The checked-in Fly configs use https://bizbox.fly.dev as an example BIZBOX_PUBLIC_URL. Change that value in fly.toml or fly.private.toml if your real Fly hostname or custom domain differs.

On first boot, Fly mounts the persistent /paperclip volume as root-owned storage. The Docker entrypoint fixes ownership before starting Bizbox as the unprivileged node user; if you see EACCES errors under /paperclip, rebuild and redeploy so the latest entrypoint is in the image.

Useful operational commands:

make status
make logs
make ssh
make secrets

If Fly Postgres logs show Out of memory: Killed process ... (postgres) or repeated backend 'bk_db' has no server available, scale the database app before debugging application routes:

fly scale memory 2048 --app bizbox-db
fly deploy --app bizbox

Use your actual database app name, for example bizbox-db-test.

Required Fly runtime settings:

Variable	Purpose
BETTER_AUTH_SECRET	Required signing secret for authenticated mode. make fly-secrets generates this with openssl rand -hex 32.
BIZBOX_PUBLIC_URL	Canonical public URL, for example https://bizbox.fly.dev. Used for auth callbacks, invite links, and hostname allowlisting.
BIZBOX_DEPLOYMENT_MODE	Must be authenticated on Fly.
BIZBOX_DEPLOYMENT_EXPOSURE	Keep private unless you are intentionally configuring a public authenticated deployment.
BIZBOX_HOME	Persistent Bizbox data root. In Fly this is mounted at /paperclip.
BIZBOX_MIGRATION_AUTO_APPLY	Applies pending migrations at startup. Set to true for this single-app Fly deployment.
BIZBOX_DB_POOL_MAX	Runtime Postgres pool cap per app process. The Fly config defaults to 3 to avoid exhausting small Fly Postgres instances.
BIZBOX_DB_IDLE_TIMEOUT_SECONDS	Closes idle runtime Postgres connections after this many seconds. The Fly config defaults to 30.
BIZBOX_DB_CONNECT_TIMEOUT_SECONDS	Fails unreachable Postgres connection attempts quickly. The Fly config defaults to 5.

If you change the Fly app name, update both APP in Makefile and app/BIZBOX_PUBLIC_URL in fly.toml, then rerun make fly-secrets.

FAQ

How is Bizbox different from agents like OpenClaw or Claude Code? Bizbox uses those agents. It orchestrates them into a company — with org charts, budgets, goals, governance, and accountability.

Why should I use Bizbox instead of just pointing my OpenClaw to Asana or Trello? Agent orchestration has subtleties in how you coordinate who has work checked out, how to maintain sessions, monitoring costs, establishing governance - Bizbox does this for you.

Do agents run continuously? By default, agents run on scheduled heartbeats and event-based triggers (task assignment, @-mentions). You can also hook in continuous agents like OpenClaw. You bring your agent and Bizbox coordinates.

Is Bizbox suitable for enterprise use? Yes. Bizbox is a fork of paperclipai/paperclip with an emphasis on enterprise usability: self-hosted deployment, complete data isolation, immutable audit trails, multi-user governance, cost controls, and on-premise compatibility. The architecture keeps your data on your own infrastructure with no external orchestration service required.

Development

pnpm dev              # Full dev (API + UI, watch mode)
pnpm dev:once         # Full dev without file watching
pnpm dev:server       # Server only
pnpm build            # Build all
pnpm typecheck        # Type checking
pnpm test             # Cheap default test run (Vitest only)
pnpm verify:agent     # Standard Codex/agent hand-off check
pnpm verify:full      # Full pre-handoff verification
pnpm test:watch       # Vitest watch mode
pnpm test:e2e         # Playwright browser suite
pnpm db:generate      # Generate DB migration
pnpm db:migrate       # Apply migrations

pnpm test does not run Playwright. Browser suites stay separate and are typically run only when working on those flows or in CI.

See doc/DEVELOPING.md for the full development guide.

Roadmap

• ✅ Plugin system (e.g. add a knowledge base, custom tracing, queues, etc)
• ✅ Get OpenClaw / claw-style agent employees
• ✅ companies.sh - import and export entire organizations
• ✅ Easy AGENTS.md configurations
• ✅ Skills Manager
• ✅ Scheduled Routines
• ✅ Better Budgeting
• ✅ Agent Reviews and Approvals
• ✅ Multiple Human Users
• ✅ Deliverables & Work Products
• ⚪ Memory / Knowledge
• ⚪ Cloud / Sandbox agents (e.g. Cursor / e2b agents)
• ⚪ Enterprise SSO / SAML integration
• ⚪ Role-based access control (RBAC) for board operators
• ⚪ Kubernetes / Helm deployment

This is the short roadmap preview. See the full roadmap in ROADMAP.md.

Community & Plugins

Find plugins and extensions on GitHub.

Telemetry

Bizbox collects anonymous usage telemetry to help improve the product. No personal information, issue content, prompts, file paths, or secrets are ever collected. Private repository references are hashed with a per-install salt before being sent.

Telemetry is enabled by default and can be disabled with any of the following:

Method	How
Environment variable	BIZBOX_TELEMETRY_DISABLED=1
Standard convention	DO_NOT_TRACK=1
CI environments	Automatically disabled when CI=true
Config file	Set telemetry.enabled: false in your Bizbox config

Contributing

We welcome contributions. See the contributing guide for details.

Enterprise

Bizbox is designed for enterprise deployments:

• Self-hosted — deploy on your own infrastructure, no vendor lock-in
• Docker-ready — production Docker Compose with PostgreSQL, health checks, and volume persistence
• On-premise compatible — no required external network dependencies
• Audit trails — every agent action and board decision is logged
• Data isolation — complete company-level data separation in a single deployment
• Cost controls — hard budget limits prevent runaway agent spend
• Governance — approval gates for agent hires, strategy, and board overrides

See doc/DOCKER.md for enterprise deployment guides.

Community

• GitHub Issues — bugs and feature requests
• GitHub Discussions — ideas and RFC

About

Bizbox is a fork of paperclipai/paperclip, rebranded and extended for enterprise use by Zest.

License

MIT © 2026 Bizbox / Zest

---

_{Open source under MIT. Built for enterprises that want to run companies, not babysit agents.}