SandboxAQ lets any AI agent join your Slack or Teams channel

SandboxAQ has released Switch, open-source software designed to integrate any AI agent directly into team collaboration channels. The platform connects agents into shared workspaces within tools like Slack, Microsoft Teams, and Discord, allowing humans and AI to collaborate with unified context and history. Through MCP, APIs, and adapters, Switch avoids vendor lock-in, enabling teams to utilize agents and models across providers. “Every team we talk to has capable agents trapped in separate tools, and a coordination cost is reducing the productivity those agents were supposed to deliver,” said Mohammed Aboul-Magd, General Manager, SandboxAQ.

Switch Enables AI Agent Integration Across Collaboration Platforms

This functionality addresses a common challenge; SandboxAQ notes that many teams currently have capable agents operating in isolation, which diminishes productivity. Developers can connect existing agents built with frameworks such as Claude Code, Google ADK, LangChain, and OpenAI, fostering interoperability and flexibility. The software deploys quickly, with both Quick Start and Team editions available, and features an extensible architecture suitable for enterprise-level implementation. A key design element is the concept of “rooms,” or channels that centralize work and maintain context as team members and agents join, leave, or transfer tasks.

The development team itself leveraged Switch during its creation, collaborating across engineering, design, and marketing within these shared rooms throughout the development and launch phases. This internal use case demonstrates the software’s practical application in real-world scenarios, such as incident response where specialized agents can be instantly integrated into a live channel with pre-existing timeline information.

Similarly, marketing teams can utilize research, copy, and design agents within a campaign room, building upon existing assets, while legal teams can expedite deal reviews with agents already familiar with the full context. Mohammed Aboul-Magd, General Manager, SandboxAQ, explained that “Switch pulls agents into the channels where teams already work, where knowledge compounds instead of being relearned with every task.

If you can message a colleague, you can put an agent to work.” Switch is available on GitHub under the Apache 2.0 license with the Commons Clause, allowing for free use within enterprises and for the creation of free products incorporating the software. It extends SandboxAQ’s Flint AI portfolio, building upon the Flint CLI, a tool developers use to scan and evaluate individual agents, to facilitate collaborative workflows involving both people and AI, the company says. The software preserves context and history within rooms, ensuring that team output builds upon previous contributions rather than resetting with each handoff, a feature designed to maximize efficiency and knowledge retention.

Every team we talk to has capable agents trapped in separate tools, and a coordination tax eating the productivity those agents were supposed to deliver.

Mohammed Aboul-Magd, General Manager, SandboxAQ
Stay current

See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.

Avatar of Ivy Delaney

Ivy Delaney

Ivy Delaney has been working with neural networks and machine learning since the mid-nineties, back when a couple of hidden layers and a long afternoon of training counted as ambitious. She has watched the field go from academic curiosity to the thing quietly running underneath everything, and she brings that long view to quantum computing. For Quantum Zeitgeist she covers the ground where the two fields meet. That means quantum machine learning and the variational algorithms it leans on, and it also means the less glamorous but more interesting story of classical machine learning already doing real work inside quantum machines, decoding error-correcting codes, calibrating noisy hardware and learning the error models that simulators depend on. She writes about the hardware those algorithms have to run on too, and about the post-quantum cryptography scramble that the same hardware has set off. Her stories typically start with the paper, whether that is peer-reviewed work, conference proceedings or an arXiv preprint, with the source linked so you can hold a claim up against the research it came from. She is unimpressed by benchmarks that will not say what they beat, and by demonstrations that only work in the press release.

Latest Posts by Ivy Delaney: