awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
CopilotKit avatar

CopilotKit/OpenBot

0
View on GitHub↗
10,393 stars·1,296 forks·TypeScript·MIT·0 viewsopen.bot↗

OpenBot

OpenBot is a self-hosted workspace where teams run AI bots as coworkers. Each bot gets its own isolated computer with a browser, workspace files and tools. Agents from any framework can be connected over the AG-UI protocol, and built-in bots can run alongside them. People converse with bots one-to-one or in groups, watch the bot's screen and the commands it runs, and can take over the browser when an agent hits a login wall or a two-factor prompt.

Governance is central to the project. Every browser, file and tool action passes through a policy gateway that checks permissions and writes an audit record. Administrators can set approval rules, grant tool access per bot, write fail-closed allow and deny policies, and review permitted, refused and failed actions in an audit log. Agents are blocked from reaching private network hosts unless an address is explicitly allowlisted, and secrets are kept out of conversation transcripts. Sign-in is available through Google, Microsoft, Okta, or company SAML and OIDC providers routed by email domain. A single-user mode is also available for a fresh install.

Beyond the core, bots can hold standing roles and goals, run scheduled routines, learn workflows from demonstrations and feedback, use reusable skills, remember facts from connected apps, and answer questions from company documents with cited sources. Conversations can be reached from desktop, web, mobile, Slack and Microsoft Teams. Model requests can be routed to OpenAI, Claude, Gemini or open models, including through a custom gateway, or run fully locally.

Deployment options include a desktop app that bundles a Docker-based backend, a single container image, Docker Compose, and a Helm chart for Kubernetes clusters. The application is an MIT-licensed template that can be forked, rebranded and modified. Some intelligence and memory services are separately licensed platform components.

Features

  • Coding Agent Workspaces - Self-hosted workspace that runs agents from any framework over the AG-UI protocol, each in its own isolated browser and file environment.
  • Agent Connectivity Interfaces - Connects agents from any framework over the AG-UI protocol so they share the same server, runtime and interface.
  • Agent Sandboxed Computers - Gives each person an isolated container with a workspace folder and browser profile per agent, optionally sandboxed further with gVisor.
  • AI Governance Tools - Team platform where AI bots work as coworkers under sign-in, per-tool permissions, approval steps and audit logs on every action.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI
  • Agent Governance - Applies sign-in, roles, tool permissions, approvals and audit logging across the whole workspace so every agent follows the same policies.
  • AI Project Workspaces - Gives teams AI bots with their own computers, browsers, memory and skills to handle ongoing research, browsing and document work alongside people.
  • Agent Computer Environments - Gives each bot its own isolated computer with a browser, workspace files and tools, managed on Docker or Kubernetes.
  • Compute Isolation - Provisions a dedicated computer with its own browser, workspace files, and tools for each person or isolated bot through a central supervisor.
  • Bot Conversations with Screen Viewing - Lets people open one-to-one or group conversations with an AI coworker while watching its screen and seeing exactly which commands it ran.
  • Action Policy Gateways - Passes every browser, file and tool action through a policy gateway that checks permissions and writes an audit record.
  • Per-Agent Desktop Isolation - Gives each person a dedicated computer for bots with separate browser profiles and files, optionally hardened with gVisor and browser sandboxing.
  • Agent Action Guardrails - Combines company sign-in, tool permissions, approval steps and audit records so teams can control and review what agents may do.
  • Agent Endpoint Registration - Lets administrators register any AG-UI agent endpoint as a bot, verifying the address and keeping authentication headers write-only.
  • Agent Takeover Controls - Hands the browser to a person when an agent hits a login wall or two-factor prompt, records the handover, and refuses agent actions meanwhile.
  • Multi-Surface Bot Access - Reaches the same bots from desktop, web, mobile, Slack and Teams while letting people watch bot activity and take over the browser.
  • Bot Observation and Handover - Shows a bot using its browser and tools, presents interactive results for review, and lets a person step in when the bot needs help.
  • Connected App Tools - Links users' own apps to their bots so the bots can read from and act on those services on their behalf.
  • Agent Tool Integrations - Registers and connects agents from any framework over the AG-UI protocol and MCP tools so different agent stacks share one workspace.
  • Unsafe Tool Call Blockers - Routes each browser and file tool call through a server gateway that checks policy and logs it before the computer acts.
  • Model Routing Gateways - Routes OpenAI, Anthropic and Google model requests through a custom gateway or proxy by overriding each provider's base URL.
  • Cross-Surface Workspace Access - Reaches one shared workspace from desktop, web, mobile, Slack and Teams, adapting conversations, results and approvals to each surface.
  • Kubernetes Agent Backends - Hosts agent runtimes and per-user bot computers, installable on Kubernetes through a Helm chart or as a single Docker container.
  • Agent Workspace Deployments - Runs the shared agent workspace on a laptop, a single container or a Kubernetes cluster so data stays in the team's infrastructure.
  • Private Network Egress Allowances - Lets a deployment reach a private-network agent by allowlisting its single address instead of opening the whole network.
  • Agent Action Approval Rules - Lets users and teams set which bot actions proceed without asking first, using personal rules, team rules and shared account requests.
  • Domain-Restricted Team Sign-Ins - Enables login through Google, Microsoft, Okta or company SAML and OIDC providers, several at once, routed by email domain.
  • Agent Action Policies - Restricts what a bot's computer may do through a JSON policy and controls which private hosts bots and agents may reach.
  • OpenID Connect Company Sign-Ins - Connects Google, Microsoft, Okta, SAML and OIDC identity providers routed by email domain and assigns administrator roles.
  • Default Deny Policies - Defines allow and deny rules over tools, pages, commands and files, evaluating deny first and refusing everything when a policy is missing or broken.
  • Sign-In Mode Configuration - Lets administrators register company SAML or OIDC providers at runtime, routed by email domain, and manage signed-in people.
  • Agent Activity Displays - Shows the agent's screen alongside a live list of commands run and files read or saved, with saved file contents kept hidden.
  • Bot Activity Feeds - Records the commands and file actions each bot performs, including permitted, refused and failed ones, so administrators can review what was done and why.

    • Star history

    Star history chart for copilotkit/openbotStar history chart for copilotkit/openbot

    How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

    Frequently asked questions

    What does copilotkit/openbot do?

    OpenBot is a self-hosted workspace where teams run AI bots as coworkers. Each bot gets its own isolated computer with a browser, workspace files and tools. Agents from any framework can be connected over the AG-UI protocol, and built-in bots can run alongside them. People converse with bots one-to-one or in groups, watch the bot's screen and the commands it runs, and can take over the browser when an agent hits a login wall or a two-factor prompt.

    What are the main features of copilotkit/openbot?

    The main features of copilotkit/openbot are: Coding Agent Workspaces, Agent Connectivity Interfaces, Agent Sandboxed Computers, AI Governance Tools, Agent Governance, AI Project Workspaces, Agent Computer Environments, Compute Isolation.

    Which projects share features with copilotkit/openbot?

    Projects with overlapping indexed features include: elie222/rakazo — Rakazo is a self-hosted, open-source platform of persistent AI bots. Each bot keeps its own conversations, memory,… kirodotdev/kirocrew — KiroCrew is a persistent workspace for AI-assisted development built around an autonomous coding agent. Tasks are… microsoft/agent-governance-toolkit — The agent-governance-toolkit is a framework for enforcing security policies, managing zero-trust identities, and… copilotkit/opendots — OpenDots is a platform for hosting long-lived AI agents, called dots, that keep running over time instead of living… omnigent-ai/omnigent — Omnigent is a policy-governed orchestrator for coding agents. It runs several agent harnesses, such as Claude Code and… builderz-labs/mission-control — Mission Control is a self-hosted web dashboard that acts as a control plane for autonomous AI coding agents. It lets…

    Projects sharing features with OpenBot

    These projects share indexed features with OpenBot. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
    • elie222/rakazoelie222 avatar

      elie222/rakazo

      7,458View on GitHub↗

      Rakazo is a self-hosted, open-source platform of persistent AI bots. Each bot keeps its own conversations, memory, scheduled routines, and history, and can hand work to peer bots or short-lived helper agents. Bots get isolated shared or private computers with a browser, terminal, files, and a graphical desktop, and users can watch a bot's screen and take control of it when needed. The same bots are reachable from web, desktop, and mobile apps, and from messaging channels such as iMessage, Slack, Telegram, and WhatsApp. The model and the computer are both your choice. You can connect your own

      TypeScriptai-agentschatgptdocker
      View on GitHub↗7,458
    • kirodotdev/kirocrewkirodotdev avatar

      kirodotdev/KiroCrew

      4,830View on GitHub↗

      KiroCrew is a persistent workspace for AI-assisted development built around an autonomous coding agent. Tasks are planned, carried out, checked against the spec, and retried on failure, with state saved to checkpoints so long-running work can resume after a restart. One shared runtime serves the desktop app, web dashboard, command line, and chat channels such as Slack and Discord, so the same ongoing work is visible from each of them. The agent improves with use: user corrections and task failures are turned into lasting lessons, applied globally or limited to a single repository, and repeate

      Pythonagentagentic-aiagents
      View on GitHub↗4,830
    • microsoft/agent-governance-toolkitmicrosoft avatar

      microsoft/agent-governance-toolkit

      6,535View on GitHub↗

      The agent-governance-toolkit is a framework for enforcing security policies, managing zero-trust identities, and sandboxing the execution of autonomous AI agents. It provides a governance layer designed to control the behavior of agents through the use of a security policy engine, cryptographic identity management, and a runtime execution sandbox. The project distinguishes itself through a multi-tier privilege ring system and a cryptographic identity mesh that secures communication between autonomous entities. It implements a decay-based trust scoring mechanism to track entity reliability and

      Python
      View on GitHub↗6,535
    • copilotkit/opendotsCopilotKit avatar

      CopilotKit/OpenDots

      6,461View on GitHub↗

      OpenDots is a platform for hosting long-lived AI agents, called dots, that keep running over time instead of living inside a single short chat. Each agent gets its own sandboxed computer with a browser, files, and a terminal that persist across restarts, with permission checks, an action record, and the option for a human to take over. The same AI coworker carries one conversation across written chat, real-time voice calls, and Slack, and a workspace of shared spaces and nested pages sits alongside it. People stay in control of what agents do. The chat interface streams agent activity live, w

      TypeScript
      View on GitHub↗6,461
    Compare all 30 related projects→