CopilotKit

OpenBot: self-hosted alternative to Grok Bot that works with any agent and any harness (9.8k stars) Grok Bot and other hosted AI coworkers (OpenAI Dots, Meta Muse, Claude Duet) lock you into their model, their agent and their cloud. Your data and the agent’s computer all live on their side. OpenBot gives you AI coworkers you run yourself. Each Bot gets its own computer with a browser, files and a shell, and you decide what it’s allowed to touch. It’s built on AG-UI (Agent-User Interaction Protocol), so you can bring any AG-UI agent. LangGraph, CrewAI, Mastra, Claude Agent SDK, Google ADK, Pydantic AI, Agno, etc or something you built yourself, it just plugs in as a Bot. You can also pick a model for each Bot so you can use a cheap model for simple tasks and a heavier one for the hard stuff. Hosted tools by big companies don’t let you do that. What you get: - A computer per Bot you can watch live, and take over when it hits a login or 2FA screen - Policy rules plus an audit log of what was allowed or refused, and why - Connect third-party tools via MCP (Drive, Notion, GitHub, etc) with more through Composio - Talk to your Bots wherever you work: web chat, voice calls, Slack, Teams or plain text messages - Built-in routines that run on a schedule - Desktop app (sign in with Claude or Codex), mobile and a browser extension - 13 ready coworkers to start from (ticket triage, release notes, expense claims and more) Memory and threads run on CopilotKit Intelligence, which you can self-host entirely. It also has something called Automatic Learning - when you correct a Bot, it picks up the pattern and turns it into a skill.md you can review, edit and ship. The Bot uses it the next time without redeploying. Self-hosting is one Docker Compose, a single image or a Helm chart, with your data in your own Postgres. 9.8k stars, 1.2k forks, 50 contributors and 604 merged PRs in about 2.5 months. TypeScript, MIT licensed. CopilotKit/OpenBot

OpenBot — Open-source AI coworkers that each get a computer of their own: a browser, files and tools, with every action decided before it happens and recorded after.... Link: CopilotKit/OpenBot #OpenSource #GitHub #DevTools

OpenDuet is an open source, self-hostable version of Claude Duet that works with any agent harness. THE PROACTIVE AGENT ERA JUST WENT OPEN SOURCE. WHY RENT IT WHEN YOU CAN OWN IT? CopilotKit/OpenDuet

OpenMuse: A personal agent with a browser, terminal, files, and work that keeps going (6.0k stars) What it does All text below is quoted directly from the project README. >A personal agent with a browser, terminal, files, and work that keeps going built with CopilotKit and AG-UI. Features: - Chat: CopilotKit headless chat with streamed AG-UI events, mailbox search and reading, send/stop in one input pill, a visible follow-up queue, retained drafts, delegated tasks, and inline email, browser, PDF, plan, and finance cards. - Agent computer: Persistent browser profiles and takeover console; optional isolated Linux terminal, saved command receipts, editable workspace files, and PDF transfer. - Activity: Durable task plans, progress, input requests, pause/resume/cancel/retry, approvals, and saved receipts. SQL leases recover interrupted work. - Goals & Tracking: Goals and milestones; recurring public-page checks for changes, text availability, or USD price thresholds, with deduplicated alerts and failure backoff. - Finance: Import transaction CSV to create a spending summary with categories, transactions, and a savings-goal action. - Gmail & Calendar: Google OAuth adapters, complete mail threads, drafts/attachments, calendar discovery, and reviewed event creation/update/deletion. Live credentials required. - Personal context: Editable name, tone, avatar, and memories. Background-update preferences and durable in-app notifications. >OpenMuse is a personal-agent application with an agent computer, visible work, and rich results. It runs its own server, task worker, and browser worker. Repository CopilotKit/openmuse Repository stats |Metric|Value| |:-|:-| |Stars|5,970| |Forks|860| |Last release|None| |Languages|TypeScript 99.4%, Other 0.6%| |Contributors|36| |AI contributor accounts (Claude / Codex / Cursor / Copilot)|None| |AI co-authored commits|Claude: 7, Codex: 2| Main contributor |Field|Value| |:-|:-| |Name|Jerel Velarde| |GitHub|@jerelvelarde| |Commits to this repo|39 (top contributor)| |Bio|Staff Applied AI Engineer @CopilotKit AI agents & generative UI Pr| |Followers|228| ────────────────────────────── 📌 If you want to browse more repos like this, we've organized them by category and rankings here: bestgithubrepositories.com 🗂️ We're also archiving repos where possible, so useful projects don't disappear when they're removed from GitHub.

OpenBot: The AI assistant your company can actually own (8.4k stars) What it does All text below is quoted directly from the project README. >Open-source AI coworkers that each get a computer of their own: a browser, files and tools, with every action decided before it happens and recorded after. Bring any AG-UI agent. Features: - A computer per person: the supervisor gives each person their own container, with a folder and browser profile for each of their Bots. - A shell, not just a browser: a Bot can run a command in its workspace, install what it needs, and process a file it saved. - The gateway is the only way in: it resolves the target from a server-held snapshot, evaluates the policy, writes the audit row, and only then calls the computer. - Take the wheel: a Bot that hits a login wall or a 2FA prompt asks for help. - Secrets never enter the transcript: the trail records that a secret was requested and how long it was, not what it said. - Bring your own agent: any AG-UI endpoint is a Bot, on a framework or hand-written. - An audit trail you can read: /admin/audit lists what was permitted, what was refused and what failed, and every refusal carries the rule that caused it. - Credentials encrypted at rest: stored through /admin/credentials, never returned by an API, and redacted from audit events. >cp .env.example .env Repository CopilotKit/OpenBot Repository stats |Metric|Value| |:-|:-| |Stars|8,436| |Forks|1.1k| |Last release|v0.2.0 (2026-10-12)| |Languages|TypeScript 89.3%, Rust 8.3%, Python 1.3%| |Contributors|47| |AI contributor accounts (Claude / Codex / Cursor / Copilot)|@claude (77 commits)| |AI co-authored commits|None| Main contributor |Field|Value| |:-|:-| |Name|David McKay| |GitHub|@davidmckayv| |Commits to this repo|431 (top contributor)| |Company / Location|Seattle, WA| |Followers|66| ────────────────────────────── 📌 If you want to browse more repos like this, we've organized them by category and rankings here: bestgithubrepositories.com 🗂️ We're also archiving repos where possible, so useful projects don't disappear when they're removed from GitHub.

OpenBot (GitHub Trending #3) The best open source Grok Bot / AI coworker available. Repo (9.1k ⭐️): CopilotKit/OpenBot t.co

Create beautiful agent UX Give your agent every UI component it might need in just 2 lines of code. It will respond with text when it's relevant, and generative UI when it's smarter. Customizable and open source: CopilotKit/CopilotKit t.co

今日のGitHubトレンド CopilotKit/OpenBot 自社インフラで運用し、自由にカスタマイズできるAIアシスタントの構築用テンプレートです。 AG-UIを通じて任意のエージェントを導入でき、個別ブラウザ環境の提供や動作への介入が可能です。 テキスト以外のコンポーネント形式での回答にも対応しています。 製品としての提供ではなく、クローンして自社専用に作り上げることを目的としています。

OpenBot's AG-UI Runtime: Per-Agent Browser Isolation and Audit-First Orchestration OpenBot is a template repository that implements AG-UI (Agent Governance UI) as a runtime layer for multi-agent systems. Each agent gets its own isolated browser session, file system, and tool set. Every action passes through an approval gate before execution and gets recorded afterward. The project positions itself as "the AI assistant your company can actually own" because it runs on your infrastructure and ships as source code, not SaaS. The repository hit 8,015 stars and trending #5 in TypeScript. It supports any AG-UI-compatible agent, including LangGraph and MCP-based systems. The architecture shows how to build governance into the UI layer rather than bolting it on after the fact. What AG-UI Means in Practice AG-UI is a specification for agent interfaces that prioritize human oversight. OpenBot implements it as a runtime that treats agents as coworkers with dedicated channels. Each coworker arrives with: - A browser session that persists logins and cookies - A file system scoped to that agent - A tool set you grant explicitly - A screen you can watch in real time The "decided before it happens and recorded after" claim means every tool call hits an approval gate. The human can approve, reject, or take over the session. After execution, the action and result get logged to an audit trail. This is different from post-hoc logging or LLM-level guardrails. The approval gate sits between the agent's decision and the tool invocation. The agent proposes, the runtime enforces, the human decides. Architecture: Isolation Without Containers OpenBot runs on a single machine (your laptop or a dedicated server) and isolates agents at the process and session level, not with VMs or containers. Browser isolation uses Playwright's browser contexts. Each agent gets its own context with separate cookies, local storage, and cache. Contexts share a browser process but maintain separate state. This is lighter than spinning up full browser instances per agent. File system isolation uses workspace directories. Each agent writes to a scoped folder. The runtime enforces path boundaries so agents cannot read or write outside their workspace. No chroot or namespace isolation, just path validation at the tool layer. Tool boundaries are enforced by the AG-UI runtime. When you configure an agent, you grant specific tools. The runtime checks the tool name against the agent's permission set before invoking. If an agent tries to call a tool it does not have, the runtime rejects the call before it reaches the tool implementation. | Component | Isolation Mechanism | Trade-off | |-----------|---------------------|-----------| | Browser | Playwright contexts | Lighter than full instances, but shared process means one crash can affect all | | File system | Workspace directories + path validation | Simple to deploy, but relies on correct path handling in tools | | Tool permissions | Runtime allowlist check | Centralized enforcement, but requires explicit grants for every tool | | Audit log | Append-only file per agent | Easy to inspect, but no built-in rotation or encryption | Approval Gate Data Model The approval gate is a state machine per tool call. When an agent decides to invoke a tool, the runtime creates an approval request with: - Agent ID - Tool name - Tool arguments (serialized) - Timestamp - Request ID The request enters a pending state. The UI shows the request to the human. The human can: - Approve: Runtime invokes the tool and records the result - Reject: Runtime returns an error to the agent - Take over: Runtime pauses the agent and hands control to the human After the human decides, the runtime appends the decision and result to the agent's audit log. The log is a JSON Lines file with one entry per action. Each entry includes the request, decision, result, and execution time. This model works for synchronous tool calls. Asynchronous or background tasks require a different pattern (polling or webhooks) that OpenBot does not currently implement. Deployment Shape OpenBot is a monorepo with workspaces for the runtime, UI, and example agents. You clone the repository, configure your agents in the examples/ directory, and run the whole stack locally. Local setup: # Clone and install git clone https://github.com/CopilotKit/OpenBot.git cd OpenBot npm install # Configure environment cp .env.example .env # Edit .env with your LLM API keys and agent config # Start runtime and UI npm run dev The runtime starts a Node.js server that manages agent processes and browser contexts. The UI is a Next.js app that connects to the runtime via WebSocket. Each agent gets a channel in the UI where you see its screen (a live view of the browser context) and approve tool calls. Production deployment is not documented in detail. The repository is in beta and assumes you will adapt it. Likely paths: - Run on a dedicated server with systemd or Docker Compose - Use a reverse proxy (nginx, Caddy) to expose the UI - Store audit logs in a database instead of files - Add authentication and multi-tenancy if you need multiple users The "no hosted version" claim means CopilotKit does not operate OpenBot as a service. You run it yourself. Integration with Agent Frameworks OpenBot supports "any AG-UI agent." In practice, this means agents that expose their tool calls and accept approval responses. The repository mentions LangGraph and MCP in topics, but the integration layer is not fully documented. LangGraph integration likely works by wrapping LangGraph's tool invocation with an approval callback. Before LangGraph calls a tool, it hits the AG-UI runtime's approval endpoint. The runtime blocks until the human decides. After approval, LangGraph proceeds with the tool call. MCP (Model Context Protocol) integration would follow a similar pattern. MCP defines how agents request context and tools. The AG-UI runtime intercepts MCP tool requests, enforces permissions, and logs results. The key is that the agent framework must support interruptible tool calls. If the framework invokes tools directly without a hook for approval, you cannot plug in AG-UI without modifying the framework. Observability and Failure Modes Observability is file-based. Each agent writes to: - An audit log (JSON Lines) - A screen recording (optional, via Playwright) - A console log (stdout/stderr from the agent process) The UI shows live browser views and recent tool calls. There is no built-in metrics system (Prometheus, Datadog) or distributed tracing. You would add that yourself if you need it. Failure modes: - Browser crash: If the shared browser process crashes, all agents lose their contexts. Playwright will restart the browser, but agents lose session state (cookies, local storage). - Agent process crash: The runtime detects the crash and can restart the agent, but in-flight tool calls are lost. - Approval timeout: If the human does not respond to an approval request, the agent blocks indefinitely. You need a timeout policy (auto-reject after N seconds). - Audit log corruption: If the log file is corrupted or deleted, you lose the audit trail. No built-in backup or replication. - Tool permission bypass: If a tool does not validate paths or permissions correctly, an agent could escape its workspace. The runtime enforces tool allowlists, but the tool implementation must enforce boundaries. Code Example: Approval Gate Hook Here is a simplified version of how an agent might integrate with the AG-UI approval gate: // Agent calls this before invoking a tool async function requestToolApproval( agentId: string, toolName: string, args: Record<string, unknown> ): Promise<{ approved: boolean; reason?: string }> { const requestId = crypto.randomUUID(); // Send approval request to runtime const response = await fetch('/api/approve', { method: 'POST', headers: { 'Content-Type': 'application/json' }, body: JSON.stringify({ agentId, toolName, args, requestId }) }); // Block until human decides const decision = await response.json(); // Log decision to audit trail await appendAuditLog(agentId, { requestId, toolName, args, decision: decision.approved ? 'approved' : 'rejected', reason: decision.reason, timestamp: new Date().toISOString() }); return decision; } // Agent uses the hook before tool invocation async function invokeTool(toolName: string, args: Record<string, unknown>) { const approval = await requestToolApproval(this.agentId, toolName, args); if (!approval.approved) { throw new Error(`Tool call rejected: ${approval.reason}`); } // Proceed with tool invocation const result = await this.tools[toolName](args); // Log result await appendAuditLog(this.agentId, { toolName, result, timestamp: new Date().toISOString() }); return result; } This pattern works for synchronous tools. Asynchronous tools need a different approach (polling or webhooks) to avoid blocking the agent indefinitely. Technical Verdict Use OpenBot when: - You need multi-agent orchestration with human oversight baked into the runtime - You want per-agent browser isolation without spinning up full VMs - You can run the stack on your own infrastructure and adapt the template to your needs - You need audit trails for compliance or debugging - You are building an internal tool where you control the agent frameworks Avoid OpenBot when: - You need a hosted, multi-tenant SaaS solution - Your agents require container-level isolation for security - You cannot tolerate single-process failure modes (browser crash affects all agents) - You need production-grade observability (metrics, tracing, alerting) out of the box - Your agent frameworks do not support interruptible tool calls OpenBot is a template, not a product. You will need to adapt it. The AG-UI pattern it demonstrates (approval gates at the runtime layer, per-agent isolation, audit-first design) is the real contribution. If you need that pattern, OpenBot gives you a working starting point. If you need a finished product, look elsewhere. Source Links - OpenBot GitHub Repository - CopilotKit OpenBot Landing Page

CopilotKit's OpenDots supports a persistent computer per agent, chat, calls and Slack, and pauses non-read-only MCP tools for an approval card. What it is, and where your data goes: CopilotKit/OpenDots #AIAgents #TypeScript #MCP