How-Tos & Practical Guides

Stripe’s developer relations team published something unusually candid on May 14, 2026: a detailed breakdown of a dozen real experiments they ran trying to get AI agents to use the Stripe API correctly. Not an announcement, not a product pitch — an honest post-mortem on what worked, what failed, and what it changed about how they think about agent design. The headline conclusion, embedded in the post title: You can’t whisper at an AI agent. ...

The NSA’s Artificial Intelligence Security Center published formal MCP security guidance on May 20, 2026 — flagging authentication gaps, overprivileged tools, and insecure context handling as systemic risks across MCP deployments. Two days later, an open-source tool landed that addresses exactly those concerns in a runnable, automated format. @hailbytes/mcp-security-scanner — published by David McHale (@hailbytes) — scans MCP server configurations for the three most common security issues that the NSA guidance calls out: overprivileged tools, missing authentication on transports, and prompt injection surface in unsanitized tool descriptions. ...

The agentic AI development stack is getting a serious upgrade. CopilotKit — the open-source framework for building AI copilots and agents into applications — has introduced three new tools that together address a critical gap in the ecosystem: how do you build, test, and debug the UI layer of multi-agent systems? The three new tools are AG-UI, AIMock, and Pathfinder. Each addresses a distinct phase of the agentic development lifecycle. Together, they represent a coherent opinionated stack for agentic UI development in 2026. ...

OpenClaw v2026.5.20 dropped on May 21st, 2026 — and it’s a release worth understanding in detail. This update touches five distinct areas: Discord voice sessions, the Doctor security linter, xAI OAuth for headless environments, a broader security overhaul of skill execution, and a new Policy plugin system. Plus a handful of fixes and smaller improvements. Here’s a complete walkthrough of what’s new, what changed, and what you need to know before upgrading. ...

Anthropic Silently Fixed a Second Claude Code Sandbox Bypass — 130 Versions, No CVE, No Advisory If you’ve been running Claude Code between version 2.0.24 and 2.1.89, your network sandbox wasn’t doing what you thought it was. For approximately 5.5 months — across roughly 130 versions — a null-byte hostname injection vulnerability allowed an attacker to bypass the network sandbox’s allowlist entirely, enabling arbitrary outbound connections and potential exfiltration of credentials, source code, or any sensitive data accessible from your sandbox environment. ...

How to Scale OpenClaw Agents on Kubernetes with ACP and acpx Running one OpenClaw agent on your laptop is easy. Running a hundred of them reliably, in parallel, across a production cluster — that’s a different problem entirely. At an AI Engineer event, Onur Solmaz, OpenClaw’s core maintainer at Hugging Face, showed exactly how to do it. The talk, titled “Scaling Agents on Kubernetes with acpx and ACP,” is the most authoritative take on production-grade OpenClaw infrastructure available right now — and it introduces a Go-based Kubernetes operator that makes horizontal agent scaling significantly more tractable. ...

How to Set Up Cloudflare CASB with Claude Compliance API for Enterprise Agent Governance Enterprise security teams finally have a native, agentless way to monitor what Claude is doing across their organization — and it doesn’t require touching endpoints, installing agents, or routing all traffic through a proxy. On May 21, 2026, Cloudflare announced that its Cloud Access Security Broker (CASB) now supports Anthropic’s Claude Compliance API. The integration gives security and compliance teams out-of-band visibility into Claude Enterprise and Claude Platform activity, directly in the Cloudflare dashboard. ...

JetBrains Koog Hits 1.0: Building Your First Production AI Agent in Kotlin If you’ve ever tried to build a reliable AI agent on the JVM and ended up with a spaghetti mess of LLM calls, retry logic, and unexplainable failures — JetBrains just dropped your solution. At KotlinConf'26 in Munich, Koog reached stable 1.0, and it came with a real-world production story: Mercedes-Benz is using it to power vehicle maintenance scheduling agents at dealers worldwide. ...

A year ago, your organization had a dozen AI agents. Today you have hundreds, maybe thousands. Every team moved fast, built their own setup, and now you have a sprawling estate of agents — each with different authentication schemes, different tool access, different logging configurations. Then the CISO asks: “Which agents are accessing customer PII?” That’s the problem Databricks addressed in a May 20 engineering blog post, and the solution they’re presenting is Unity Catalog extended to govern the entire agent stack — not just data, but models, tools, and the MCP servers that connect agents to enterprise systems. ...

Your AI agent has a ticket queue full of infrastructure requests. It has read access to your runbooks, write access to your deployment pipelines, and the ability to execute changes against live systems. It also reads Jira tickets, wiki pages, and Slack transcripts to decide what to do next. That combination — broad access plus natural-language reasoning from untrusted inputs — is the attack surface security teams need to be thinking about right now. ...