# Introducing the SageOx Agent Toolkit > Source: > Published: 2026-08-31 00:00:00+00:00 # Introducing the SageOx Agent Toolkit One Agent. One Memory. Everywhere Your Team Works. TL;DR: We're launching an open-source, Buzz-compatible toolkit to bring SageOx's team memory to your hosted agents. At SageOx, our mission is to build the hivemind layer that makes AI agents team-aware. When Block released Buzz, we were thrilled because it finally created the [agent-forward workspace](/blog/welcome-buzz) we've been waiting for. We realized there was a gap, though: while Buzz provides the place for humans and agents to work together, those agents start every session as new hires with no shared history. It is helpful to distinguish between history and memory. Buzz creates a perfect record of events, which is powerful. However, a raw event log is just history. It is not a memory. Real team knowledge requires distillation. Actionable insights still require figuring out what is relevant, what was decided, why, and by whom. Our toolkit adds that layer of provenance and makes your agents instantly as smart as the rest of your team. Agents built on this toolkit cite who made which decision and learn from actual team intent, rather than just guessing from a pile of raw data. Beyond the agentic surface provided by Buzz, we believe the true potential lies in augmenting agents with team memory. We're unlocking this by integrating SageOx's team context directly into the Buzz ecosystem. We are releasing this toolkit to help teams move past the "cold start" problem. Now, your agents don't just act in your workspace; they arrive with the shared context of your team's decisions, conventions, and discussions. This toolkit unlocks an entirely new way of working: treating agents as true team members rather than transient scripts. It's how we've successfully deployed internal agent fleets to handle complex, ongoing workflows. We did not just build this in theory. I recently joined SageOx and during my [work trial](/blog/reflections-on-my-work-trial), I connected our toolkit to a Buzz workspace. We were able to ask our agents, "What is the team working on?" and get an immediate summary that spanned coding sessions, meetings, and shared chats. Buzz + SageOx is the perfect combination and a massive unlock. We could take full advantage of Buzz's purpose-built workspace while seamlessly integrating deep, shared memory of the entire team - memory built across humans, agents, in code-sessions, conversations, and surfaces outside of Buzz. We are taking the lessons from that trial and putting them directly into this toolkit. Our goal is to make team context follow you wherever you work. Buzz is just the starting point, and we're rapidly expanding support to Slack, Telegram, and beyond to ensure your agents operate seamlessly across the tools your team already uses. We're in alpha and we want to build this with you. If you're a developer looking to deploy team-aware agents, or just want to see how this fits into your workflow, please reach out. Let's build something together. You can start building your agents [here](https://github.com/sageox/agent-toolkit) today. At SageOx, we believe the future of work is a blend of human-human, human-agent, and agent-agent collaboration. We're building the connective tissue to make that happen, and we'd love for you to join us.