When Anthropic announced on April 4 that Claude Pro and Max subscribers could no longer use their flat-rate plans through third-party agent frameworks, the developer community felt it immediately. More than 135,000 OpenClaw instances were running when the restriction took effect. Now the creator of OpenClaw — who joined OpenAI in February — has gone public with a sharp response, calling the decision a “betrayal of open-source developers.” What Anthropic Actually Did The policy change is straightforward: Claude’s subscription tiers were designed around conversational use. A human opens a chat window, types a query, reads a response. Agentic frameworks operate differently — a single OpenClaw instance running autonomously through a day can consume the equivalent of $1,000 to $5,000 in API costs. ...

Anthropic has shipped Ultraplan — a new Claude Code feature that runs your most complex planning tasks in a remote cloud container, powered by Opus 4.6 in plan mode, while you continue working locally. When the plan is ready, you get a diagram-rich document you can review, approve, or hand back to Claude Code for execution. The short version: you give Claude Code a hard problem, Ultraplan disappears into the cloud for up to 30 minutes to think about it properly, and you don’t have to wait. ...

When Anthropic’s April 4 restriction on using Claude subscriptions with third-party tools took effect, 135,000 active OpenClaw instances stopped working as their operators expected. The financial math was stark: users who built workflows on Claude Max at $200/month were now facing API costs of $1,000–$5,000/month to maintain the same usage levels. The reaction from OpenClaw’s creator — now working at OpenAI — was direct and public. ‘A Betrayal of Open-Source Developers’ In a statement to The Next Web published April 4, the creator of OpenClaw described Anthropic’s restriction as “a betrayal of open-source developers who built entire platforms on the promise that subscription tiers would remain available to automation.” ...

The Allen Institute for AI (Ai2) has released MolmoWeb — an open-source web browser agent that controls a browser through visual understanding alone. No DOM access, no custom instrumentation. Just screenshots, a model, and a set of learned actions: click, type, scroll, navigate. It comes in two sizes: 4B and 8B parameters. The 8B model scores 78.2% on WebVoyager, 42.3% on DeepShop, and 49.5% on TailBench — benchmarks where it outperforms GPT-4o-based agents. The weights, training code, and dataset (MolmoWebMix) are all Apache 2.0. ...

The freshest item from today’s Claude Code release landed just five hours ago: agents can now declare initialPrompt in their YAML frontmatter, and Claude Code will auto-submit that prompt as the first turn when the subagent is invoked. No user input required, no manual kickstart — the agent launches and immediately has context to work with. It’s a small feature. It’s also the kind of small feature that quietly reshapes how multi-agent workflows get built. ...

Anthropic’s accidental Claude Code source leak, first reported last week, has had a consequence that security researchers were quietly warning about: someone used the exposed code to find a real, critical vulnerability. This is distinct from the Vidar malware campaign that exploited brand confusion around the leak (also covered here previously). That was opportunistic social engineering — attackers leveraging the story of the leak to distribute malware. What SecurityWeek is reporting now is different: researchers with access to Claude Code’s 600,000-line codebase — exposed via npm source maps — used that access to conduct legitimate offensive security research and found a critical functional vulnerability. ...

If you’ve ever watched a coding agent confidently write Gemini API code that was deprecated six months ago, Google has something for you. Two new tools launched this week from Google’s developer team — Gemini API Docs MCP and Gemini API Developer Skills — and together they do something impressively concrete: push coding agent accuracy on Gemini API tasks from roughly 60% to 96%, according to Google’s own evals. That’s not a marginal improvement. That’s the difference between an agent that’s useful and one that’s reliably useful. ...

CVE-2026-33579 is a critical privilege escalation vulnerability in OpenClaw (CVSS 8.1–9.8) that allowed anyone with operator.pairing scope — the lowest permission level — to silently grant themselves full admin access. It was patched in v2026.3.28, but the exploit leaves no obvious trace. Security experts recommend that any OpenClaw instance running a pre-patch version should be treated as potentially compromised, even without visible evidence of breach. This checklist walks you through the full audit process. ...

A Silicon Valley fintech startup has done what many have theorized and few have actually shipped: replaced its entire software development team with AI agents built on OpenClaw and Claude Code. JustPaid — which makes an AI-powered platform for automating financial operations like billing and invoicing — has deployed seven fully autonomous AI coding agents. In just one month, those agents built ten major features. Each one, says CTO Vinay Pinnaka, would have taken JustPaid’s human developers a full month to ship. ...

If you’re running a self-hosted OpenClaw instance and haven’t patched in the last week, stop what you’re doing. Security researchers are using a phrase that should make any sysadmin’s stomach drop: “assume compromise.” That’s not alarmism. It’s a measured response to CVE-2026-33579 — a critical privilege escalation vulnerability in OpenClaw that was patched earlier this week, but not before potentially exposing thousands of installations to silent, undetectable admin takeover. What Is CVE-2026-33579? The vulnerability affects all versions of OpenClaw prior to v2026.3.28. Its CVSS score ranges from 8.1 to 9.8 depending on the metric used — squarely in the “critical” band. ...