OrgoAI/bops

Bops: a team of AI bots that run your business ops, each with its own computer, email and phone number. bops.bot

Bops – AI‑powered business‑operations bots for macOS

What it is – Bops is a macOS‑only desktop application (built with Electron + Next.js) that lets you run a team of AI “bots” to automate everyday business tasks. The bots execute their work on real cloud computers provided by Orgo, while you interact with them through a chat‑style UI inside the app or via Slack, Telegram, Discord, email, SMS or voice calls.

Key capabilities

  • Real‑computer execution – Each bot runs on an Orgo cloud VM (4 CPU, 16 GB RAM, multi‑screen Chrome, LibreOffice, GIMP, Inkscape). You can watch the screen live or take control.
  • Team of bots – A main bot (Boppy) coordinates specialist bots. Bots keep conversation threads, remember context per workspace, and can hand work to each other.
  • App & channel integration – Chat with bots in the Bops app, or from Slack, Telegram, Discord, email (a dedicated *.bops.bot address) and phone (dedicated numbers on Pro/Max plans).
  • Secure app access – Connected apps (≈1,400 services) are stored in a “Vault” backed by the Mac Keychain; passwords and 2FA never leave the device.
  • Human‑in‑the‑loop – Bots ask for approval before sending emails, creating invoices, or making payments. A “Just do it” mode can bypass prompts for low‑risk actions.
  • Data look‑ups – On‑demand enrichment of companies, people, funding rounds, etc., for a small per‑lookup fee.
  • Automation – Routines, reminders, screen watches, and the ability to draw/edit images directly in chat.

Plans & pricing

Plan Price Cloud computers AI credit Phone/email
Free $0 1 VM, 10 h/month $5 one‑time none
Pro $20 / mo 1 always‑on VM $20 / mo 1 phone, 1 email
Max $200 / mo up to 3 VMs $200 / mo up to 5 phones, 5 emails

All plans include unlimited bots sharing the allocated VM(s). Additional AI credit can be purchased $5‑$1 000.

Self‑hosting – The repo contains everything needed to run Bops yourself on a Mac. Set BOPS_SELF_HOSTED=1 and provide your own API keys (OpenAI, Orgo, and optional services like TypeSafe, Honcho, Composio, etc.). The server runs locally on port 3210; a small relay (edge/) can expose webhooks for Slack, SMS, calls, etc., via Tailscale or a public URL.

How it works under the hood

  • The Electron front‑end launches a Next.js server (app/, lib/server/).
  • Bot VMs are created from an Orgo template (orgo/, vm/).
  • Tasks on the VMs use OpenAI’s Computer tool (Responses API). Tasks on the local Mac use OpenAI’s Agents API via the Codex CLI.
  • State is stored locally (~/Library/Application Support/Bops/server/.data) or in Postgres for the hosted cloud version.
  • Optional integrations (Slack, Telegram, Discord, email, phone) are wired through third‑party services (Composio, AgentMail, AgentPhone, TypeSafe, Honcho).

Getting started

  1. Download the signed DMG from <bops.bot> (Apple‑silicon macOS required).
  2. Install, sign in (Google, email, or Orgo).
  3. The app creates a default bot “Boppy”.
  4. Invite the bot to your Slack/Telegram/Discord or start a chat in the app and give it tasks.

Development & contribution

  • Run npm run setup to install dependencies and create .env.local.
  • npm run app launches the desktop client and server.
  • Build releases with scripts/release.sh (requires Apple Developer ID, notarization credentials, and the private Orgo relay binary).
  • Contribute via standard GitHub PR workflow; linting, TypeScript checks, and code‑gen steps are documented.

License – “FSL‑1.1‑ALv2” (free for use, self‑host, and modification; commercial hosting of a competing service is prohibited). After two years each version is relicensed under Apache 2.0.

Bottom line – Bops is a genuine, open‑source AI‑agent platform that lets non‑technical users automate business workflows on real cloud computers, with strong integration options and an optional self‑hosted deployment path.

Related

  • Project
    MicrosoftLearning/mslearn-ai-agents

    A repository of hands-on exercises to learn how to build AI agents on Microsoft Azure. You'll practice common tasks and need an Azure subscription with access to generative AI models.

  • Project
    Floe-Labs/floe-guard

    floe‑guard is a Python/TypeScript library that tracks the real USD cost of every LLM, speech‑to‑text, text‑to‑speech, and other AI‑vendor call in‑process, and can hard‑stop a call before it would exceed a user‑defined dollar ceiling. It ships with adapters for OpenAI, Anthropic, Gemini, LangChain, CrewAI, LiteLLM, Vercel AI SDK, and voice platforms (Pipecat, LiveKit, Vapi, Retell). A bundled public‑price cost map provides offline pricing; you can override it with your own rate card. Budgets can be persisted across processes via SQLite, and an optional hosted Floe service adds a Coverage Score and 7‑day history. The core API is `BudgetGuard(limit_usd)`, with `check()` (pre‑call) and `record()` (post‑call) methods, plus atomic `reserve_tool`/`settle_tool` for paid tools.

  • Project
    beevibe-ai/beevibe

    An agent-native operating system that creates a shared workspace for humans and AI agents to collaborate, preventing the loss of shared intelligence across a team.

  • Project
    jacke-pp/open-yoki

    open-yoki is a self-contained AI livestream host that runs on one server, with a cloned voice, digital avatar, chat interaction, memory, and autonomous posting. It integrates with Binance Square and TwinOasis, and is fully customizable.