Supermemory Opens Company Brain, Its Slack AI Bot With Thousands of Users
Supermemory open-sourced its discontinued Company Brain product, a Slack teammate with persistent memory that runs on your own Cloudflare account.
- Supermemory open-sourced Company Brain, a Slack teammate previously sold as a paid SaaS product
- One-click deploy to your own Cloudflare account, needs only a supermemory key and one model provider key
- Permissions modeled as a graph so the bot only reads what the asker can already see
- Ships with MCP tool integrations for GitHub, Linear, Notion, Google Workspace and more
- Runs on Cloudflare free tier, capped at 50 outbound calls per request; Workers Paid ($5/mo) removes the ceiling
- TypeScript stack on Workers, Durable Objects and D1, Apache 2.0 licensed
Supermemory has released the Company Brain repository under Apache 2.0. According to the company, the Slack assistant previously ran as a paid product with thousands of users. Supermemory discontinued that service several weeks ago and rebuilt the application for deployment in a customer’s Cloudflare account.
Apache 2.0 allows teams to inspect, modify, use, and redistribute the code subject to the license terms. Teams control the application runtime and Cloudflare resources, while Supermemory’s hosted API stores the bot’s memory. Model requests also go to the selected provider, so deployments retain external dependencies for memory and inference.
From channel memory to action
The README opens with an incident-response example: an engineer asks whether production is down, and the bot replies without being mentioned. Its response includes the affected endpoint, recent error count, teammate investigating the problem, and likely triggering deployment. The example shows the project’s defining behavior: it monitors conversations, retrieves relevant context, and can respond proactively.
- Persistent memory: Captures decisions, project context, and ownership from the Slack channels available to it.
- Tool actions: Connects to GitHub, Linear, Notion, Google Workspace, and other services through Model Context Protocol, a standard interface for exposing tools to AI agents.
- Sandboxed execution: Clones repositories, runs shell or Python tasks, and returns artifacts such as charts, CSV files, and PDFs in Slack.
- Proactive replies: Contributes without a direct mention, with chattiness controls at the organization and channel levels.
This story is for Pro members
You've reached the end of the free preview. Upgrade to AlphaSignal Pro to read the full article - and everything else behind the paywall.
Comments
