AI For That

openTag

Visit

Open source AI coworker for Slack and Lark

openTag introduction

openTag is an open source AI coworker that lives in Slack and Lark. You mention it in a thread with a task, it runs a coding agent on your own computer, and it replies in the same thread.

Website:
opentag.build
πŸ’» Code & DB Assistantfree
X
openTag screenshot

openTag overview

openTag connects team chat to AI agents running on your own machine. You create an agent, prepare your computer with a connection command, and connect your messaging app (Slack or Lark). When you mention openTag in a thread, the openTag app hands the request to your computer, where a small background program runs your coding agent with your own subscription or API key, and the reply is sent back to the thread. It supports agent runtimes including Codex, Claude Code, and Pi, and is described as having no AI provider lock-in. Shared team memory is handled through Context Tree, a portable, file-based, git-backed memory layer hosted in your GitHub organization.

openTag features

  • Open source under Apache 2.0
  • Works in Slack and Lark
  • Runs coding agents on your own computer
  • Supports Codex, Claude Code, and Pi agent runtimes
  • Bring your own subscription or API key
  • No AI provider lock-in
  • Customisable agent profiles
  • Team mode: invite your agent to team channels
  • Agents can hand work to each other and check on each other's progress
  • Context Tree shared team memory: file-based, git-backed, hosted in your GitHub organization
  • Connects to over 1000 popular services including Slack, Lark, GitHub, Notion, Linear, Google Drive, Jira, and Figma
  • Local first on Mac and Linux, with Docker self-hosting

Questions about openTag

openTag pros

  • Open source under Apache 2.0
  • No AI provider lock-in; works with Claude or Codex plans or an API key
  • Agents run locally on your own computer
  • Supports Slack and Lark
  • Team mode lets one agent serve a channel
  • Context Tree team memory is file-based, git-backed, and hosted in your own GitHub organization
  • By default, openTag's servers don't store conversations, code, or memory
  • Free to use, paying only for your own AI use

openTag limitations

  • Requires a Mac or Linux computer with Claude, Codex, or Pi installed
  • Requires a Slack or Lark workspace
  • Teams and Discord support is only on the roadmap
  • Subscription plans should be used in accordance with each provider's terms of use, and an API key is recommended for shared workspaces

openTag use cases

  • Answering questions about project status in a team channel
  • Pulling together team discussions and project notes
  • Triaging repository issues and opening pull requests
  • Reviewing pull requests
  • Sharing team memory across AI agents
  • Delegating tasks from chat to an agent on your machine

Who openTag is for

  • Engineering teams
  • Software engineers
  • Open-source maintainers
  • Teams using Slack or Lark
  • Teams wanting to use their own AI subscriptions or API keys

openTag pricing

openTag is free to use. You pay for your own AI use with your existing provider subscription.

Alternatives to openTag