OpenBot

What is OpenBot?

OpenBot is a CopilotKit project for running AI assistants that your company owns. It looks like a chat assistant, but it deploys on your infrastructure and you can change any part of it. Each agent gets its own channel, like a teammate, with a thread per agent and slash-command skills limited to what that agent has been granted. Agents answer from company documents in Google Drive, OneDrive and Notion with sources attached, and retrieval fails closed so an agent only sees files the asking person may already open. When no API exists, an agent can sign in and operate a tool through its own container with a separate browser, while you watch the screen and take over at any point. Newer releases add bot-to-bot delegation and scheduled routines. You bring any agent that speaks the AG-UI protocol, such as LangGraph, Mastra or PydanticAI, and use your own model keys. It is MIT licensed and offered as a template to clone and adapt rather than a hosted product, so there is no OpenBot price; the CopilotKit pricing page covers a separate service.

Use cases

• Answer company questions from Drive and OneDrive documents
• Give each AI coworker its own role and skills
• Automate tools that have no usable API
• Run scheduled routines with agents
• Triage an inbox and draft replies
• Deploy an internal agent workspace on your own servers

Who it's for

• Engineering teams
• IT and platform teams
• Companies building internal agents
• AI developers

Why use OpenBot

• Keeps documents, vectors, permissions and conversations in your own Postgres, so nothing leaves your environment unless you configure it.
• Respects existing file permissions, returning nothing rather than guessing when access is unclear.
• Records every tool call and outcome, and lets a person watch or take over an agent's browser session.
• Avoids lock-in to one agent framework because any agent that speaks AG-UI can be registered.
• Treats agents as named roles with limited skills and sources, which makes them easier to reason about and review.

OpenBot features

• One channel and thread per agent
• Skills exposed as slash commands, limited to granted ones
• Answers from Google Drive, OneDrive and Notion with cited sources
• Fail-closed retrieval that respects source permissions
• Per-agent container with its own browser and files
• Live view of the agent's screen with option to take over
• Bring any AG-UI agent, including LangGraph, Mastra and PydanticAI
• Bot-to-bot delegation and scheduled routines
• Gated MCP tool calls with a per-tool allowlist and recorded outcomes
• Admin-held, encrypted model provider keys with no bundled model
• Docker Compose demo and Helm path to production

OpenBot integrations

• Google Drive
• OneDrive
• Notion
• Slack
• Microsoft Teams
• LangGraph
• Mastra
• PydanticAI
• MCP

How to get started with OpenBot

1. Clone the OpenBot repository and start the demo with Docker Compose, or deploy with Helm for production.
2. Have an admin add model provider keys and connect sources such as Google Drive, OneDrive or Notion.
3. Register agents that speak AG-UI and grant each only the skills and sources its job needs.
4. Message an agent in its channel, or @mention another, and watch it work, including any browser actions.
5. Review the recorded actions, add scheduled routines, and customize the code for your company.

OpenBot limitations

• It is a template, not a product: there is no hosted version to sign up for
• Nothing is published as a package, so you clone the repository and customize it
• Adapting it for your company requires engineering work
• Google Drive access is read-only in v0.0.5, while Notion supports reading and writing
• The built-in agent is unavailable until an admin sets up a model provider

OpenBot pricing

OpenBot is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

OpenBot alternatives

The main OpenBot alternatives listed on Debuly are Open WebUI, LibreChat, Dify, Glean, OpenClaw and AnythingLLM. The table compares pricing and platforms; order is A–Z.

1. 

   AnythingLLM

   Open-source AI app for chatting with your documents and running agents, on desktop, in Docker or as a hosted team workspace.

   Chat assistantsAutomation and agentsDeveloper platforms
   Open sourceVisit AnythingLLM (official site)
2. 

   Dify

   Platform for building LLM apps, agents and RAG workflows on Dify Cloud or your own infrastructure, with a free Sandbox tier.

   Developer platformsAutomation and agentsChat assistants
   FreemiumVisit Dify (official site)
3. 

   Glean

   Enterprise AI platform that searches company apps, answers questions with permissions respected, and runs agents on your own data.

   ProductivityResearch and searchAutomation and agents
   Contact salesVisit Glean (official site)
4. 

   LibreChat

   Free open-source AI chat platform that puts many model providers, agents, code execution and MCP tools in one self-hostable app.

   Chat assistantsDeveloper platformsAutomation and agents
   Open sourceVisit LibreChat (official site)
5. 

   Open WebUI

   Self-hosted AI chat platform for Ollama and OpenAI-compatible models with RAG, agents, tools, notes, channels and role-based access.

   Chat assistantsDeveloper platformsAutomation and agents
   Open sourceVisit Open WebUI (official site)
6. 

   OpenClaw

   Open-source personal AI assistant that runs on your own devices and takes actions for you through WhatsApp, Telegram, Slack, Discord and other chat apps.

   Automation and agentsChat assistantsProductivity
   Open sourceVisit OpenClaw (official site)

OpenBot FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: none collected yet, so a monogram is shown.
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.