MCP App StoreLe meilleur de l'écosystème IA et MCP, sélectionné chaque jour.
Sources
OpenAI and Paradigm launched EVMbench, a specialized benchmark for evaluating AI agents' ability to find and fix high-severity smart contract vulnerabilities. This provides a critical tool for improving agent reliability in secure coding and blockchain environments.
GPT-5.2 has successfully proposed a new formula for gluon amplitude in theoretical physics, which was subsequently verified by academic collaborators. This marks a significant milestone in LLMs contributing to frontier scientific discovery.
OpenAI introduced Lockdown Mode and Elevated Risk labels to protect organizations from prompt injection and data exfiltration. These features are essential for enterprises deploying AI in high-security environments.
OpenAI released GABRIEL, an open-source toolkit that uses GPT to convert qualitative text and images into quantitative data. This allows social scientists to analyze large-scale research datasets with high precision.
OpenAI details the engineering behind their real-time access system, integrating rate limits and usage tracking to scale Sora and Codex. Essential reading for developers building high-throughput AI infrastructure.
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.
OpenAI launches GPT-5.3-Codex-Spark, a real-time coding model offering 15x faster generation and 128k context. Now in research preview for ChatGPT Pro users.
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.
An exploration of how Codex is being utilized to build robust agentic workflows. Focuses on the engineering patterns required for an agent-first development environment.
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.
OpenAI's GPT-5 combined with Ginkgo Bioworks' automation reduced cell-free protein synthesis costs by 40%. This demonstrates the power of closed-loop AI experimentation in biotechnology.
OpenAI Frontier is a new enterprise platform specifically for managing AI agents. It provides critical infrastructure for shared context, permissions, and governance at scale.
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.
The GPT-5.3-Codex system card reveals the most capable agentic coding model yet. It integrates frontier coding performance with advanced reasoning and professional knowledge.
OpenAI introduces GPT-5.3-Codex, a native agent combining frontier coding performance with general reasoning. It is designed to handle complex, long-horizon technical tasks and real-world software engineering.
OpenAI details the Codex App Server, a bidirectional JSON-RPC API that enables embedding Codex agents. The system supports streaming progress, tool execution, and diff-based approvals for technical work.
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 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.
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.