The latest from the AI and MCP ecosystem, curated daily.
OpenAI demonstrates a bidirectional Codex + Figma workflow: generate a running UI with Codex, bring it into Figma for design refinement, then sync changes back to Codex for implementation. This closes the design-to-code loop in a way that keeps both designers and developers in their preferred tools. An interesting workflow for teams bridging design and AI-assisted development.
Anthropic describes integrations between Cowork and finance plugins that streamline secure data access workflows for financial applications. For developers, the post highlights plugin patterns and permission models that make it easier to connect private datasets to agent workflows while retaining controls. This is relevant to teams building regulated-agent integrations where secure plugin architectures are required.
Google introduced Opal — a new tool for building dynamic, visual agentic workflows with Gemini, letting developers design multi-step agent pipelines with conditional branching, tool calls, and state management through a drag-and-drop interface. Targets developers who want to prototype agent architectures quickly before building them in code.
Anthropic launched Cowork, a collaborative workspace feature that lets teams work alongside Claude together in real time, plus a plugin system for connecting Claude to internal enterprise tools. This moves Claude beyond individual productivity into shared team workflows and custom integrations. A significant product expansion for enterprise buyers.
Anthropic outlines how AI-assisted migration tools reduce the cost and risk of modernising legacy COBOL systems, using code understanding and automated transformation to accelerate porting. This matters for enterprises still running critical COBOL workloads — developer teams can use these patterns to prioritize modernization with lower upfront engineering effort. Practical case studies indicate measurable reductions in time and cost for common migration tasks.
OpenAI details how to use Codex for long-horizon tasks — multi-step jobs that run autonomously without constant user input. The post covers structuring tasks for reliable execution, using hosted shell and background mode, and managing state across extended agent runs. Essential patterns for anyone building production Codex workflows.
GGML and llama.cpp — the foundational libraries behind local LLM inference — are joining Hugging Face to ensure their long-term development and maintenance. This brings the most widely-used local inference stack under HF's open-source umbrella, with commitments to continued compatibility and community governance. A significant consolidation in the local AI ecosystem.
IBM Research and UC Berkeley released IT-Bench and MAST — a benchmark suite and failure taxonomy for enterprise AI agents in real IT environments. The research identifies the root causes of enterprise agent failures across categories like tool misuse, context loss, and ambiguity handling. Practical reference for teams debugging why their production agents underperform in enterprise settings.
Hugging Face demonstrates using Codex and Claude agent skills to generate custom CUDA kernels — making GPU-level optimisation accessible to developers without deep CUDA expertise. The post shows how agent skills can encode specialist knowledge (GPU programming patterns) and apply it on demand. A compelling use case for agent skills beyond typical software engineering tasks.
OpenEnv is a new evaluation framework that tests tool-using agents in real-world execution environments rather than sandboxed simulations — catching failure modes that controlled benchmarks miss. The post covers the framework design, evaluation methodology, and results across several agent models. Useful for any team building agents that interact with real APIs or filesystems.
OpenAI shares practical patterns for building agents that do real sustained work — combining agent skills, hosted shell execution, and server-side context compaction in the Responses API. Covers how to keep long-running agents from hitting context limits and how compaction lets them work across sessions. Useful for anyone building beyond simple single-turn agents.
Transformers.js v4 is now available on NPM, bringing Hugging Face's in-browser and Node.js inference library to a major new release with improved model support, better performance, and a cleaner API. Developers can now run transformer models directly in JavaScript without a Python backend. A big upgrade for web developers building AI-powered browser applications.
Anthropic engineers built a working C compiler by running multiple Claude instances in parallel, each responsible for different compiler stages, with a coordinator orchestrating the workflow. The experiment demonstrates real multi-agent software engineering at a non-trivial scale and surfaces practical lessons about task decomposition, inter-agent communication, and error recovery. A compelling proof-of-concept for parallel agent coding systems.
Hugging Face launched Community Evals — an open framework for community-run model evaluations that anyone can contribute to, challenging the opacity of existing leaderboards. The goal is to build a more trustworthy, diverse set of model benchmarks driven by real use cases rather than optimised for leaderboard gaming. Important infrastructure for the open-source AI evaluation ecosystem.
OpenAI distills 15 hard-won lessons from building ChatGPT Apps — covering UX patterns, capability design, handling ambiguity, and how they packaged the learnings into a Codex Skill to bootstrap new apps 10x faster. Practical and opinionated guide for anyone building on the ChatGPT Apps SDK.
Hugging Face reflects on one year since DeepSeek disrupted the AI landscape, examining how open-source AI has evolved in response — from the proliferation of capable open-weight models to new training approaches and the emerging AI+ ecosystem. The third in a series, focusing on where the global open-source AI movement goes next.
LinkedIn's AI team shares a practical retrospective on training GPT-OSS with agentic reinforcement learning — covering the infrastructure challenges, reward design decisions, and lessons learned from scaling RL training to production-grade agent models. One of the more candid engineering accounts of what it actually takes to do agentic RL at scale.
Google introduced Agentic Vision capabilities in Gemini 3 Flash — enabling agents to see, understand, and act on visual content from screens, documents, and images as part of multi-step workflows. Brings computer-vision-powered tool use to a fast, cost-efficient model tier. A key capability for agents that need to process visual context alongside text.
MCP Apps launched as the first official MCP extension, enabling tools to return interactive UI components — dashboards, forms, visualisations, multi-step workflows — that render directly in the conversation. Built as an open standard in collaboration with OpenAI and the MCP-UI community. A landmark moment for the MCP ecosystem that expands what MCP servers can deliver beyond plain text.
The MCP Core Maintainer team announces changes for 2026: Inna Harper and Basil Hosmer are stepping back, with new maintainers joining to keep the project moving. Also covers what's ahead for MCP governance as the protocol moves under the Agentic AI Foundation. Good context for anyone tracking MCP's organisational trajectory.
