OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Workflow Agents

Openbot

Openbot is an MIT-licensed, self-hostable AI coworker template that connects AG-UI agents to isolated browsers, files and selectively granted tools. A central gateway evaluates and records actions before allowing or refusing them, while PostgreSQL stores decisions and CopilotKit Intelligence stores threads.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Openbot — OpenViST project profile in Workflow Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Openbot when an organization wants to assemble and own an AI coworker environment around its own infrastructure, agent stack and granted tools rather than consume a hosted assistant.

Who it is for

  • Teams building company-owned AI coworkers.
  • Developers integrating AG-UI-compatible agent stacks.
  • Organizations that need browser, file or MCP access mediated by explicit tool grants and recorded actions.
  • Builders prepared to operate a Docker, Bun and PostgreSQL-based self-hosted workspace.

Documented uses

  • Give an agent its own browser, logins, files and workspace.
  • Connect agents built with supported AG-UI-compatible frameworks or custom implementations.
  • Control and record actions involving browsers, files, MCP servers and interface components.
  • Adapt example coworkers to company workflows such as ticket triage, policy review, knowledge questions and renewal preparation.
Before you choose

Requirements & limits

The repository states that there is no hosted version to sign up for. It is intended to be cloned and run on the user’s infrastructure. The documented setup requires Docker, Bun 1.3+, a CopilotKit Intelligence project and license, and a model key; a free Intelligence plan is described as available.

  • Self-hosting is central: the repository says the application runs on the user’s infrastructure and uses their PostgreSQL.
  • Tool governance is built around a gateway that decides, records and then executes or refuses actions.
  • AG-UI provides the integration boundary, allowing agent stacks from several named frameworks or custom implementations.
  • Each coworker has its own browser, logins, files and workspace, which suits workflows needing separated working environments.
  • The project is alpha and described as a template, so adopters should expect to customize the example tenant and handle ongoing development changes.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the Openbot workspace

Prepare a local Openbot environment and connect it to a CopilotKit Intelligence project.

Before you start

  • Docker for PostgreSQL and the shipped Bots.
  • Bun 1.3 or later for the application and API server.
  • A fresh CopilotKit Intelligence project and license.
  • A model key; the proof-of-concept Bot uses OpenAI, while the LangGraph Bot supports OpenAI, Anthropic or Google.

Steps

  1. Create the environment file from .env.example.

  2. Install the project dependencies with Bun.

  3. Log in with the CopilotKit CLI and select the Intelligence project, as described in the repository quick start.

  4. Provide the required model and project credentials in the environment configuration.

Expected outcome

A configured local Openbot workspace connected to the required Intelligence and model services.

Guide limitations

  • The repository describes Openbot as an alpha template under active development.
  • This setup recipe does not establish that a deployment has been executed or that a particular agent workflow will operate successfully.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
2 October 2026
Original website title
CopilotKit/OpenBot · Official homepage title · retained in its source language
How we review sourcesSuggest a correction