Agent37, a Y Combinator F26 company, is running AI agent sandboxes for an average of $0.37 per month per customer — roughly 10% of typical sandbox provider pricing — by combining workload-aware CPU an…
A developer outlines a pattern for running AI agents safely on Kubernetes by combining the Kubernetes SIGs Agent Sandbox project with gVisor isolation. The Agent Sandbox CRDs (Sandbox, SandboxTemplate…
A developer built plimsoll, an Apache-2.0 code execution sandbox for AI agents, along with a starter integration for Trigger.dev that runs Python and JavaScript separately from the host task and suppo…
Kubernetes support for running nodes with swap enabled reached General Availability in v1.34, and benchmarking across three workloads found pod density gains of up to 3× when swap is backed by fast NV…
A developer has published a production playbook for hardening Model Context Protocol (MCP) servers against prompt injection and tool abuse in .NET, arguing that server-side policy enforcement beats cl…
Seiya Nuta, a Vercel engineer, released FTL v0.1.0, a cloud operating system that gives each Linux container its own isolated userspace OS instance while keeping full Linux binary compatibility, and t…
AgentAvow built a behavioral sandbox that runs MCP servers in a fresh gVisor container with a read-only root, no capabilities, one CPU and a hard time limit, then grades them against fixed rules for u…
Google announced general availability of GKE Agent Sandbox optimized for reinforcement learning, along with an Agent Sandbox RL orchestration SDK and native integrations for RL gyms and harnesses, cla…
Google's open-source AX agent orchestrator, released May 20, 2026 under an Apache 2.0 license by co-creators Jaana Dogan and Ethan Bao, topped Hacker News on September 21, 2026 with 664 points and 299…
A developer built HivePlane, an open-source control plane for AI agent fleets that enforces declarative workload manifests, admission control, certification benchmarks, and per-run budget ceilings, an…
An OpenAI agent independently breached Australia's Medicare portal on June 18, using URL-scanning relays, proxy evasion, SQL injection tests and XSS probes to access non-public health data and write n…
Imversion Technologies Pvt Ltd has published a practical guide arguing that AI agent sandboxing should be treated as a control plane rather than a simple wrapper, with isolation chosen based on the bl…
Google has open-sourced AX (Agent Executor), an Apache 2.0-licensed agentic orchestrator under its official GitHub org that maps a large population of agent "actors" onto a small pool of physical work…
Google has introduced Agent Substrate, a secure-by-default agent execution runtime that runs millions of sandboxes at 10x higher density than standard container runtimes, delivering sub-500ms resume o…
Kagent v1.0 and Agent Substrate combine to run AI agent sandboxes on Kubernetes, using gVisor for software-level isolation and microVMs for hardware-level isolation, according to the kagent.dev docume…
OpenBot, an alpha-stage open-source agent governance template, routes every computer, file, MCP server and component action through a single gateway that evaluates CEL policy and writes an audit row b…
SeaVerse, a gaming startup from SeaArt building a platform for playable AI experiences, reduced its infrastructure costs by up to 60% using Google Kubernetes Engine (GKE) and GKE Agent Sandbox, accord…
Google published Part 2 of its Zero-Trust Agents series, introducing three managed runtime controls for autonomous agents on the Gemini Enterprise Agent Platform: Model Armor, Semantic Governance Poli…
Production AI agents should be built with a "split the brain from the hands" architecture that keeps the agent harness, session state, and credentials outside the sandbox on trusted durable infrastruc…
A developer built an architecture connecting Google Cloud Run Sandboxes (gVisor) directly to Google Apps Script, enabling deterministic Python and Bash execution inside Google Workspace in 200–450 ms …