openTag for Share Team Memory
Share team memory with openTag
If your team already works in Slack or Lark and keeps decisions scattered across threads, notes, and docs, openTag gives that context somewhere an AI coworker can actually use it. You can mention openTag in a thread, delegate work to it, and have it reply in the same conversation after pulling from shared team context. That shared memory is called Context Tree: a portable, file-based, git-backed memory layer hosted in your GitHub organization.
For this use case, the important part is not just that openTag can answer in chat, but that its memory is meant to be shared across agents. Team memory lives in markdown files in a git repository you host yourself, which makes it readable, portable, and versioned. Because it is hosted in your GitHub organization, the memory stays private and under your control rather than sitting in a separate vendor-managed store.
openTag also fits teams that want the memory layer to work alongside their existing tools. The product connects to services such as GitHub, Notion, Google Drive, Linear, Jira, and Figma, and it can connect to more than 1000 popular services overall. That means the shared context openTag uses is not limited to a single chat thread; it can sit beside the systems where your team already keeps plans, notes, and project material.
The local execution model matters here as well. openTag runs agents on your own machine, and by default its servers do not store your conversations, code, or memory. The task is handled through your computer, while the reply goes back to Slack or Lark. For teams that want a shared memory layer without handing control of the underlying context to the vendor, that setup is the main appeal of openTag’s approach.
openTag’s team mode also supports channel-level collaboration. One agent can serve a channel, and anyone in that channel can mention it. The product also says agents can hand work to each other and check on each other’s progress, which makes the shared memory more useful when several people are relying on the same context rather than repeating the same background in separate requests.
If a team wants to use its own AI subscriptions, openTag is built for that as well. It supports your Claude or Codex plan, or an API key, so the shared memory layer is not tied to one AI provider. That matters for teams that want memory and coordination in one place while keeping control over the model account they use behind the scenes.