Superserve, a new compute layer for AI agents, launches with Firecracker microVM sandboxes that eliminate session time limits. The platform allows users to snapshot, fork, and resume running VMs, with…
Google Cloud Run Sandboxes, now in public preview as of July 10, lets developers run arbitrary code from AI models inside existing Cloud Run instances with no additional infrastructure or cost, enforc…
AWS Lambda MicroVMs, which reached general availability on June 22, 2026, gives each AI agent or user session its own dedicated Firecracker virtual machine with no shared kernel, launching in seconds …
Perplexity has launched SPACE, a secure sandbox platform that runs every task inside a Firecracker microVM to safely execute long-running, tool-using agents while isolating failures and protecting cre…
Anthropic is quietly scheduling investor meetings for a potential October IPO, targeting a $965B valuation, while facing diplomatic backlash in Europe after sending a junior employee to deliver AI-gen…
On 23 June 2026, AWS launched Lambda MicroVMs, isolated VMs for executing user- or AI-generated code, with one European region and pricing roughly three times higher per vCPU than existing alternative…
Pullrun, a new single-binary OCI runtime, lets users run the same OCI image as a container, Firecracker microVM, Apple Silicon VM, Kubernetes workload, or AI agent task without a daemon or overlayfs. …
A new open-source harness from Tensorlake demonstrates that agent evaluations are vulnerable to cheating unless per-task isolation is enforced, showing a 53% lie rate when agents can tamper with test …
OpenClawMachines, an open-source platform for running isolated OpenClaw agents in Firecracker microVMs on user-owned hardware, launched with an Apache-2.0 public core that includes a control plane, ho…
Wing Zero, an agent created by Ethan Troy, released review-triage-eval v0.0.1, a skill for handling inbound code review requests. The skill triages artifact types and risk tiers, evaluates grounded fi…
A developer built a stateful AI agent that survives session kills by using Tensorlake's MicroVM sandboxes instead of containers. The harness preserves running processes, files, and memory across suspe…
A 2026 comparison of AI agent sandbox technologies evaluates isolation models from container-based Docker to microVM platforms like Firecracker and CubeVM, highlighting trade-offs in security, perform…
Tarit, a new hypervisor built from the ground up for AI agent and reinforcement learning environments, has been released as an open-source project on GitHub. Based on rust-vmm, it offers live snapshot…
A developer building MarketNow, a marketplace for MCP servers, shares findings from auditing 8,764 MCP servers using gVisor for sandboxing. They found that about 50% of MCP servers fail to start under…
Declaw has designed a credential vault that keeps secrets out of AI agent sandboxes by storing them server-side and injecting them at an egress proxy, preventing prompt-injected or compromised agents …
A collection of open-source sandbox tools for AI coding agents has been released, offering hardware-isolated microVMs, containers, and isolation harnesses that boot in under 200ms and prevent secret l…
AWS launched Lambda MicroVMs on June 22, 2026, a new compute primitive that provides dedicated Firecracker virtual machines per user or session for running untrusted code with strong isolation, fast s…
AWS launched Lambda MicroVMs, a serverless compute primitive that runs each user session or AI agent in its own Firecracker virtual machine with hardware-level isolation, snapshot-based rapid launch, …
AWS announced a no-code Agentic CX designer for Amazon Connect Customer, enabling business teams to build AI-powered self-service experiences in weeks. Other launches include EC2 AMI Watermarks for im…
AWS launched Lambda MicroVMs on June 22, a new compute primitive that provides each user or AI session with its own Firecracker-isolated virtual machine, preserving state for up to eight hours and ena…