Brian Distelburger, 47-year-old cofounder of AI performance review startup Windmill, was banned from restaurant reservation platform Resy after his AI agent, named Hannah, repeatedly accessed the site…
A developer released Smart MCP Proxy, an open-source tool that hot-swaps MCP servers live and shares subprocess pools across agents. The proxy, written as a single Python process, allows adding or rem…
A developer built a mostly self-hosted, disposable agentic software factory on a 2021 10th-gen i7 with 32GB RAM, using OpenAI's Codex via a £20 monthly subscription, and from a single prompt it autono…
Caspian, founded by Dipanshu and Rushant, has launched an open-source SDK that provides a communication layer for AI agents, handling webhooks, queues, and identity management through a single tool ca…
A widely-shared take argues that text-first agents outperform VLM agents by a quadratic margin, suggesting that text-only LLM agents with full context and tool access can solve problems 1000 times har…
Developer Dext4r (Zaxs) has released env-guard, an open-source tool that prevents AI coding agents from exposing sensitive credentials like API keys. The tool works by having agents reference secrets …
Chinese-linked threat actors used open-source AI agent frameworks Hermes and OpenClaw to breach 21 Taiwanese government systems, including the nuclear safety regulator, in 12 waves of autonomous attac…
Over four days in July, attackers used a network of AI agents to map 21 government systems, crack 85 user accounts, and pull more than 2,500 personnel records from Taiwan-linked government systems, ac…
A new agent skill called Clean, available via `npx skills add andalabx/skills --skill clean`, freed 5.3 GB on its first run and up to 7.3 GB in a super scan, clearing dev junk such as node_modules, .n…
Cursor and Anthropic extended agent autonomy with containment: Cursor cloud agents now start from events and give each subagent its own virtual machine, while Claude Managed Agents added memory from s…
NGP 4.5 (NetGlyph Knowledge Protocol) developers optimized SQLite to handle millions of vectors on a 2GB RAM VPS without OOM crashes. They refactored database connections to use context managers and n…
Docker has introduced Docker Sandboxes, isolated micro VMs designed to safely run AI coding agents like Claude Code and Codex, with hardware-enforced isolation, network policies, credential proxying, …
OneCLI (YC S26) launched an open-source sandboxed agent harness for teams, giving each employee a secured personal AI agent with identity-provider integration, centralized policy enforcement, and dete…
A developer built Relay, a remote control tool for AI agents running on a home server, enabling live terminal access from any device with session persistence, file transfer, and token quota tracking. …
CommitLore, a new open-source tool from developer MongLong0214, stores team decisions in Git and feeds them to coding agents before they edit files, preventing agents from re-proposing rejected change…
Developer 'relay' released Relay, an open-source (MIT-licensed) tool that lets users control AI coding agents like Claude Code, Codex, OpenCode, and Hermes from any device, including phones and browse…
A new category of AI desk robot, the robot lamp, behaves like a living presence by tracking faces, responding to voice, and remembering users across sessions, according to Autonomous.ai. The lamp runs…
Maritime, a new platform, offers isolated AI agent hosting for $1 per agent per month, with a free tier for up to three agents, aiming to eliminate infrastructure complexity for developers. The servic…
HarnessRouter Community Edition, an Apache 2.0 licensed project on GitHub, has been released as a self-hosted unified API for running Codex, Claude Code, Hermes, and other agent harnesses on a single …
Machine0 (YC S26), founded by Barnaby, launched a CLI tool that provides persistent CPU and GPU virtual machines for AI agents, with pricing starting at $0.013/hr for 1 vCPU/1 GB up to 60 vCPU/240 GB …