SpaceXAI launches Team Bots, shared Grok Bots with team context and memory

SpaceXAI (xAI) said Monday it is launching Team Bots — Grok Bots built around a shared role or workflow so a whole team works from the same context, with plugins into apps like Salesforce, Notion, and GitHub, plus Slack handles and per-user private conversations.

While OpenAI and Anthropic dominate the agent-safety headlines, SpaceXAI is shipping a concrete shared-bot product for teams that already live in Slack, Linear, and GitHub. That is how agent software actually lands in companies.

On Monday, 28 September 2026, SpaceXAI announced Team Bots. The newsroom post is titled “Team Bots: AI coworkers that learn from your team,” and the page prints Sep 28, 2026. A Grok Bot is SpaceXAI’s assistant. A Team Bot is one of those bots built around a role or a workflow a whole team shares, so everyone can work from the same context. Context, here, means the files, instructions, and know-how the bot is given. Those lines are SpaceXAI’s.

SpaceXAI says every Team Bot brings four things together. Context is the relevant files, instructions, and skills, from a brand guide to internal documentation. Plugins let it work inside apps such as Salesforce, Notion, and GitHub. A plugin is a connection into another product. A person can connect one, or a team can set one up for everyone. Credentials let it reach a third-party API that does not have a plugin. An API is a door one piece of software uses to talk to another service. Memories keep what the bot learns, so it gets better at the role over time. Those four pieces are SpaceXAI’s.

The bot is shared. The chats are not. SpaceXAI says each person’s conversations stay private. The bot keeps separate context and memories for each user, and it still draws on the skills the team shares. Those lines are SpaceXAI’s.

Teams can also work with a Team Bot in Slack, the chat app many companies already use. Each Team Bot gets its own handle, the name you type to call it into a channel. People in that channel can ask questions, add context, and see its replies. Those lines are SpaceXAI’s.

SpaceXAI says it uses Team Bots on its own teams. In sales and customer success, every major account has a dedicated bot. The account executive, the customer-success manager, the solutions architect, and the sales leader share it. Each night the bot reviews company news, recent Gong calls, Notion docs, and Slack threads. Gong is a tool that records sales calls. Each morning it posts a briefing in the account’s Slack channel: what changed, and what each person should do next, including a draft written for that person’s role. During the day the team uses it to check a plan against the account’s data. It remembers the decisions. When people rotate on and off the account, SpaceXAI says the bot is the record that brings a new teammate up to speed. Those lines are SpaceXAI’s.

Harper, an insurance company for small businesses, is the customer example on the post. Dakotah Rice, Harper’s chief executive, is quoted saying the team used to check every customer by hand across three platforms, pull balance details, and send personal emails. They built a Team Bot in 24 hours to find customers whose policies had lapsed — coverage that had stopped — and help them reinstate it. Rice said that saved customers over $120,000 across hundreds of policies, and that the bot is now an ongoing engine. The quotation and the dollar figure are Rice’s, as SpaceXAI prints them.

In product and engineering, SpaceXAI says its Engineering Team Bot sits in a project’s Slack channel and connects to Notion, Linear, Hex, Datadog, and Cursor. Linear is a place teams file bugs. Cursor is a coding app. The bot follows product decisions, sorts bug reports, creates tickets, and launches Cloud Agents for fixes that are already well defined. A Cloud Agent is a coding agent that runs off the person’s laptop. SpaceXAI says it taught the bot the company’s shipping process: how to handle a pull-request review, when to ask the team for help, and what evidence a change needs before it counts as done. A pull request is a proposed code change waiting for review. SpaceXAI says it used this setup while building Team Bots. The bot steered a Cursor project that ran hundreds of Cloud Agents. A five-person team shipped more than 100 pull requests a day and launched Team Bots in a few weeks. Those counts are SpaceXAI’s.

In marketing, SpaceXAI says Marketing Bot has the brand guidelines, the blog posts, and the social copy. When someone shares a draft in Slack, the bot reviews it against the company’s voice and the latest messaging. For a website or search change, once the update passes review, the bot makes the change and posts a preview link in Slack for a final sign-off. SpaceXAI also quotes Angela Ferrante, head of marketing at Amplitude. She said the company is building toward Team Bots for every marketing function, to dig into market and customer information, show up in AI answers, review launches, and say what is working. That quotation is Ferrante’s, as SpaceXAI prints it.

On data, SpaceXAI says the analytics team gets dozens of one-off requests a day. Data Bot lets anyone at the company ask a question without waiting for an analyst or getting a warehouse login of their own. It uses shared, read-only credentials to query approved tables in Databricks. Read-only means it can look at those tables and cannot change them. It also connects to Datadog, Hex, and Statsig. The analytics team had spent two years writing skills for working with more than 45,000 tables, and it gave that library to the bot. The skills cover how to find the right data, how to look at feature use for fraud, how to follow SpaceXAI’s chart standards, and how to update a company-wide dashboard safely. The bot remembers corrections to its queries, so a fix from one person improves the answer the next person gets. Those lines are SpaceXAI’s.

In plain terms, a Grok Bot can now be a shared teammate. One bot for the whole squad, plugged into the tools the team already uses, that remembers the project. Each person still gets a private conversation with it.

The picture is SpaceXAI’s launch still for Team Bots. It shows Eng Bot sorting bugs in Linear and opening the matching work in GitHub. It is the product screen from the launch video.