How-Tos & Practical Guides

Nous Research shipped Hermes Agent v0.15 (versions 0.15.0 through 0.15.2) on May 28, 2026. If you’ve been tracking the Hermes project, this is a significant milestone — not just an incremental update. Two things stand out as genuinely important for practitioners: native multi-agent swarm topology and Brainworm-class prompt injection defenses. Let’s break down what changed and why it matters. The Architecture Overhaul Before getting to features, it’s worth noting the scale of the refactor. The run_agent.py core was reduced by 76% — from a sprawling monolith to a leaner 3,821-line codebase. This isn’t cosmetic. In software with a large contributor community (the v0.15 release incorporated hundreds of community PRs), a 76% reduction in core complexity means significantly easier contribution, maintenance, and debugging. ...

A post from the Obelisk engineering blog — “SQLite is All You Need for Durable Workflows” — hit the Hacker News front page this week with 392 upvotes and 211 comments. That’s exceptional signal, not algorithmic noise. When 200+ engineers are actively debating a technical architecture claim on HN, it’s worth understanding why. The thesis is deceptively simple: for a large class of durable workflow systems, especially AI agent runtimes, you don’t need Temporal, Redis, or external state stores. SQLite is sufficient. ...

Anthropic shipped Claude Opus 4.8 on May 28, 2026 — just six weeks after Opus 4.7. If you’re reading a press release summary, you’ll hear about benchmark scores. If you’re running production agents on Claude, you care about something else: what actually changes, what might break, and what you need to rebuild. This piece is the practitioner lens. Let’s get into it. What’s New in Opus 4.8 (The Parts That Matter for Agents) Effort Controls: The Reasoning Depth Dial The biggest new control surface is the effort parameter — a way to explicitly tell Claude how much reasoning to apply to a given task. ...

BadHost (CVE-2026-48710): A Critical Starlette Vulnerability Is Threatening Millions of AI Agents Disclosed: May 22–26, 2026. Patch released: May 21, 2026. This is urgent — act now. If you’re running any Python-based AI agent infrastructure — vLLM, LiteLLM, FastAPI-backed MCP servers, OpenAI-compatible proxies, or any other Starlette-dependent service — you have a critical vulnerability on your hands until you patch. Here’s everything you need to know and exactly what to do. ...

OpenClaw v2026.5.27 and v2026.5.28: What’s New and How to Use It OpenClaw shipped two significant releases this week — v2026.5.27 (stable security hardening) and v2026.5.28 (stable feature expansion with beta previews) — bringing GitHub Copilot as a first-class agent runtime, full Claude Opus 4.8 coverage, ClawHub plugin trust surfaces, and a wave of cross-channel identity fixes. Here’s what changed, why it matters, and how to make use of the new capabilities. ...

The WEF Just Told Enterprises: Stop Scaling Agents on Vibes, Start with Authorization Published May 26, 2026 — the World Economic Forum and Capgemini have released their latest joint publication: AI Agents in Action: A Playbook for Trusted Adoption, Authorization and Scaling. It’s a 60-page practitioner playbook — not an academic treatise — and its central argument is worth sitting with: the main bottleneck blocking enterprise AI agents from scaling past pilots isn’t capability. It’s authorization. ...

Anthropic Releases Claude Opus 4.8 with Dynamic Workflows — Up to 1,000 Parallel Subagents in Claude Code Anthropic just shipped something that should make every agentic AI practitioner sit up straight: Claude Opus 4.8, now available today at the same price as its predecessor, with a research-preview feature called Dynamic Workflows that lets Claude Code orchestrate hundreds — and theoretically up to 1,000 — parallel subagents on a single task. ...

AWS Rebuilds Amazon OpenSearch Serverless from the Ground Up for Agentic AI Applications AWS didn’t just update Amazon OpenSearch Serverless. They rewrote it. Announced today by Channy Yun on the AWS News Blog, the next generation of Amazon OpenSearch Serverless is a ground-up rebuild designed specifically for the traffic patterns of agentic AI workloads — and the differences from the previous version reflect how fundamentally the move to agents changes infrastructure requirements. ...

CoreWeave Closes the Training-to-Inference Gap with Unified Agentic AI Platform The “deploy it and hope it improves” era of AI agents is ending. CoreWeave just made it official. Today, CoreWeave announced a unified agentic AI platform that creates what it calls a superintelligence loop — a closed feedback cycle where production agents continuously improve from real-world data using reinforcement learning, without requiring practitioners to stitch together separate training, inference, and observability pipelines. It’s a significant architectural shift for anyone building serious agentic AI infrastructure. ...

Hackers Use LLM Agent to Move from Marimo RCE (CVE-2026-39987) to Internal Database in Four Pivots This is the attack security researchers have been warning about for two years. On May 10, 2026, it happened. Sysdig Threat Research Team (TRT) has documented what they’re characterizing as the first confirmed LLM-agent-driven intrusion chain: attackers exploited a critical remote code execution vulnerability in Marimo notebooks (CVE-2026-39987), deployed an LLM agent to autonomously navigate the victim environment, and exfiltrated an internal PostgreSQL database — all within a single hour, with no static playbook, no human-in-the-loop decision making, and just four AI-guided pivots. ...