OpenAI just launched GeneBench-Pro, a challenging research-level benchmark designed to measure how AI agents navigate ambiguity and messy datasets in computational biology across 129 tasks. Using synthetic data with known causal structures for deterministic grading, GPT-5.6 Sol scored a 28.7% pass rate—up sharply from GPT-5's sub-5% run on the original benchmark. Ten sample questions are live on Hugging Face.
Anthropic just launched Claude Sonnet 5, making it the default model across Free and Pro tiers while keeping API pricing at $2/$10 per million tokens. It closes the performance gap with Opus 4.8 on coding and complex agentic workflows, natively planning and operating tools like terminals without hand-holding.
Anthropic's Claude Code team just dropped a practical guide on loop engineering, helping developers graduate from babysitting individual prompts to building autonomous agent loops. The framework breaks loops into four distinct levels—turn-based, goal-based using /goal, time-based using /loop and /schedule, and proactive routines—while offering tricks like SKILL.md self-verification to keep token costs in check. Basically, it shows how to turn open-ended coding agents into deterministic, self-healing workers.
Anthropic just released Claude Science in beta, an all-in-one desktop workbench for researchers on macOS and Linux. The app pulls together over 60 scientific connectors across genomics, chemistry, and HPC compute, complete with a built-in reviewer agent that catches calculation errors and bad citations. It is available now for Claude paid subscribers.
ComfyUI just launched Comfy MCP in public beta, turning AI agents like Claude and Cursor into automated creative operators. Instead of manually wiring complex node spaghetti, you can simply prompt your agent in plain English to build, edit, and run production pipelines. It is free to test during beta with a Comfy account.
Cursor just launched a native iOS app in public beta for paid users, letting developers kick off cloud-based coding agents or steer local tasks straight from their phone. It supports voice prompts, lock-screen Live Activities, diff reviews, and one-tap PR merges on the go. Mobile Composer 2.5 runs are also 75% off through July 5, 2026.
Microsoft just open-sourced Resource2Skill, a framework that distills human tutorial videos, articles, and code into executable agent skills. Instead of manual prompt writing, it builds a multimodal skill library hooked up to MCP servers, enabling agents to generate real Web pages, PowerPoint decks, and Blender scenes.
HP just announced it is scaling its Frontier strategic partnership with OpenAI across the company. The decision follows eye-popping pilot results: one engineer cleared 122 PRs across 43 projects in weeks, while the security team patched monthly backlogs in a single day. Now, HP plans to roll out AI agents across device management, coding, and its 100,000-partner portal.
Einsia AI's Navers Lab just open-sourced BrowserBC, a tool that acts like an AI-native macro recorder. Instead of brittle pixel coordinates, it records your browser actions and distills them into plain-text skill cards that any web agent can read, understand, and execute autonomously.
OpenAI just launched a limited preview of its new GPT-5.6 model family, led by the flagship Sol alongside Terra and Luna. Sol introduces an ultra multi-agent mode and deeper reasoning to tackle complex coding, biology, and cybersecurity workflows. At the U.S. government's request, rollout starts with a select group of trusted partners before expanding broadly.
Runway just rolled out Agent 2.0, turning simple text prompts into full marketing briefs and multi-format campaign assets right inside its platform. It can also digest performance data to refine creative output and scale across channels and markets, moving Runway from a standalone generation tool into automated campaign operations.
Ornith just introduced Ornith-1.0, an open-source model family designed specifically for agentic coding across sizes from 9B Dense to 397B MoE. The key breakthrough is self-scaffolding: the model learns to author its own task workflows rather than relying on human-engineered harnesses, matching top frontier models on SWE-bench Verified. Weights and edge-ready 21GB GGUF variants are live on Hugging Face.
Microsoft just upgraded Copilot in Excel with repeatable financial skills, third-party data connectors, and built-in auditability. Analysts can now standardize multi-step tasks like DCF modeling via simple SKILL.md files in OneDrive and query institutional data from PitchBook, FactSet, or S&P Global directly in spreadsheets. Crucially, a new planning step outlines every cell change before Copilot executes it.
Google just started rolling out a "Select from screen" tool in Chrome 149, letting users snip text or images directly into Gemini prompts via the "+" menu. The update landed alongside new computer use capabilities for Gemini 3.5 Flash, letting AI agents navigate browsers and desktops for multi-step tasks. It basically cuts out the tedious copy-paste routine so Gemini sees what you see.
Meta just open-sourced Astryx, the internal design system that has powered over 13,000 of its apps over the past eight years. Built on React 19, it ships 150+ accessible components, ready-made themes, and a CLI designed so human engineers and AI coding agents can build from the exact same conventions.
Meta just introduced Autodata, an agentic framework that allows AI to automatically build, test, and refine high-quality training and benchmark data. Instead of static prompts, it runs a closed loop where agents evaluate failures and upgrade their generation recipe over time. Basically, it converts inference compute directly into cleaner training datasets without human babysitting.
Google just integrated computer use directly into its mainstream Gemini 3.5 Flash model, doing away with the standalone Gemini 2.5 experimental setup. Developers can now build agents that observe and act across browser, mobile, and desktop environments out of the box, backed by new prompt injection safeguards and optional approval gates. It gives Google a fast, production-ready competitor in the automated workflow race.
Figma just introduced a design agent at Config that builds custom, reusable plugins and WebGPU shaders directly from text prompts. Designers can now spin up tailored tools on the canvas without dev setups or coding skills. It effectively turns prompt writing into functional workflow automation for teams.
Nous Research just added a /learn command to Hermes Agent, letting it distill any directory, documentation URL, or recent workflow directly into a reusable skill file. It skips heavy background engines entirely, using the agent's own file and browsing tools so it works seamlessly across local and remote setups.
Perplexity just launched Computer for Counsel, a legal workspace powered by 20+ frontier AI models that links research databases, contracts, and case files directly to Perplexity Computer. It integrates sources like Midpage for case law, plugs into Microsoft 365 and Google Workspace, and automates contract triage. It is available to Enterprise and Max subscribers.
