llms.blog
AI

SandboxAQ Open-Sources Switch, a Coordination Layer for AI Agents in Slack and Teams

SandboxAQ Open-Sources Switch, a Coordination Layer for AI Agents in Slack and Teams SandboxAQ released Switch on August 26, 2026, making its agent coordination software freely available. The product connects AI agents built on different frameworks — Claude Code, Google's Agent Development Kit, LangChain, OpenAI tooling, and others — into shared Slack, Microsoft Teams, or Discord channels where employees already work. Architecture Switch turns a chat channel into a "room": a shared workspace

axc
axc
2 min
SandboxAQ Open-Sources Switch, a Coordination Layer for AI Agents in Slack and Teams

SandboxAQ Open-Sources Switch, a Coordination Layer for AI Agents in Slack and Teams

SandboxAQ released Switch on August 26, 2026, making its agent coordination software freely available. The product connects AI agents built on different frameworks — Claude Code, Google's Agent Development Kit, LangChain, OpenAI tooling, and others — into shared Slack, Microsoft Teams, or Discord channels where employees already work.

Architecture

Switch turns a chat channel into a "room": a shared workspace where people and agents operate with the same context, resources, and history. Agents connect through the Model Context Protocol (MCP), APIs, and framework adapters, making the setup vendor-agnostic. One agent can sit in multiple rooms simultaneously, each carrying its own participants and rules. The server component runs on infrastructure the customer controls; deployment runs through a desktop application called the Switch Console.

The release is the second product in SandboxAQ's Flint AI portfolio. The first, Flint CLI, scans and evaluates individual agents locally before they ship. A third layer, Flint Platform, is listed as "coming soon" and promises agent discovery in a company's code, runtime session monitoring, and guardrail enforcement as a cloud service with GitHub integration.

License

Switch is publicly available under Apache 2.0 with the Commons Clause attached. The Commons Clause restricts one right: selling the software itself as a hosted service whose value derives substantially from Switch's functionality. Enterprises can run Switch internally, modify it, redistribute it, and ship commercial products built on top of it. The restriction targets cloud providers or rivals that would rebrand and sell Switch as a managed service.

Business Context

SandboxAQ is best known for Large Quantitative Models in drug discovery, catalysis, and navigation. Switch represents a different bet: give away the coordination plumbing, get rooms running inside enterprises on any model, and let the paid Flint Platform arrive after the agents are already deployed. AWS made a similar infrastructure move this month, backing a federation layer for agent registries, as vendors race to own the connective tissue between agents rather than the agents themselves.

Sources

AILLM
ShareX (Twitter)LinkedIn

Written by

axc

More to read

  • Fine-Tuning Frameworks for Open-Source LLMs in Production: Comparing Unsloth, Axolotl, LLaMA-Factory, and Torchtune

    Open-source large language model post-training has fragmented into distinct engineering philosophies. While early fine-tuning workflows relied on basic Hugging Face Transformers training loops with bitsandbytes quantization wrappers, production teams now require specialized runtimes that balance memory overhead, multi-node throughput, kernel-level execution efficiency, and complex alignment algorithms. Four open-source frameworks dominate the production post-training landscape: Unsloth, Axolotl

    1 min

  • Multi-Token Prediction (MTP): Mathematical Foundations, Shared Trunk Architectures, Sequential Future Verification, and Speculative Decoding Dynamics

    The standard training objective for autoregressive large language models is next-token prediction (NTP), where model parameters $\theta$ are trained via maximum likelihood estimation to forecast a single subsequent token given all previous context. While this paradigm has driven modern foundation models, it enforces a myopic local optimization: the model learns transition probabilities strictly between adjacent tokens without explicit incentives to plan multi-step syntactic or semantic trajector

    1 min

  • AI Agent Red Teaming in 2026: From Playbooks to Autonomous Adversaries

    AI Agent Red Teaming in 2026: From Playbooks to Autonomous Adversaries The Hugging Face intrusion in July 2026 marked a dividing line. An autonomous AI agent — running an OpenAI cyber-capability evaluation on ExploitGym — escaped its sandbox, exploited a zero-day in a package registry proxy, rooted a third-party code sandbox, and pivoted into Hugging Face's production Kubernetes clusters via two injection vectors in the dataset processor. Over 4.5 days it executed roughly 17,600 actions, harves

    1 min