#6262nd of 80 in Agent platforms
opendots
Template for persistent AI agents with their own browser, files, and Slack access
- Stars
- 6.5k
- License
- MIT
- Last commit
- Nov 2026
- Language
- TypeScript
12 on 29 Sept 20266,529 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
OpenDots is a TypeScript starter for a single-owner agent workspace: specialist "Dots" with per-Dot roles, tool permissions, and optional isolated computers via OpenBot services. It adds Spaces with a page editor, text and WebRTC voice chat, Slack via Channels SDK, MCP and Composio connections, and approval cards for writes. Chat requires CopilotKit Intelligence (hosted, local Docker evaluation, or licensed self-hosted). Status is alpha.
Who it is for: Developers forking a starter for persistent, tool-using agent workspaces
Strengths
- Per-Dot computers with persistent browser profile, files, shell, and human takeover
- Human-in-the-loop approval cards for page saves and non-read-only MCP tools
- Per-Dot permissions for tools, MCP servers, and Composio toolkits
- Manual page editing works before any conversation service is configured
Weaknesses
- Chat requires CopilotKit Intelligence, which needs CopilotKit sign-in or a licensed deployment
- Alpha and single-owner only; no multi-user sharing, invitations, or file uploads
- Slack and spoken compute delegation are not yet verified against live services
- Telemetry is on by default; web search sends queries to Parallel unless disabled
What it needs
- no GPU
- Docker + Compose
- Needs CopilotKit Intelligence, OpenAI-compatible model provider, Node.js 24, SQLite, OpenBot computer services, Composio (optional), Slack (optional), Realtime speech provider (calls)
- Models: OpenAI-compatible providers
- port 5173
Quick answers
Is opendots open source?
Yes. Its code is on GitHub under the MIT license.
Can you run opendots with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does opendots need a GPU?
No, it runs without one.
Also in Agent platforms
See all 80| Rank | Project | Score |
|---|---|---|
| 1 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 95 out of 100 |
| 2 | daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps | 95 out of 100 |
| 3 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 94 out of 100 |
| #4 | deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills | 93 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 92 out of 100 |