はてなブックマーク

Published Nov 04, 2025 Direct tool calls consume context for each definition and result. Agents scale better by writing code to call tools instead. Here's how it works with MCP. The Model Context Protocol (MCP) is an open standard for connecting AI agents to external systems. Connecting agents to tools and data traditionally requires a custom integration for each pairing, creating fragmentation an

Have you ever asked an AI model what’s on its mind? Or to explain how it came up with its responses? Models will sometimes answer questions like these, but it’s hard to know what to make of their answers. Can AI systems really introspect—that is, can they consider their own thoughts? Or do they just make up plausible-sounding answers when they’re asked to do so? Understanding whether AI systems ca

Anthropic officially opens Tokyo office, signs Memorandum of Cooperation with the Japan AI Safety Institute This week, we opened our first Asia-Pacific office in Tokyo, a milestone in Anthropic's international expansion. Our CEO and co-founder Dario Amodei traveled to Tokyo to meet with Prime Minister Takaichi, address members of the LDP Digitization Headquarters Committee, meet customers and sign

Beyond permission prompts: making Claude Code more secure and autonomous Published Oct 20, 2025 Claude Code's new sandboxing features, a bash tool and Claude Code on the web, reduce permission prompts and increase user safety by enabling two boundaries: filesystem and network isolation. In Claude Code, Claude writes, tests, and debugs code alongside you, navigating your codebase, editing multiple

Today, we're introducing Claude Code on the web, a new way to delegate coding tasks directly from your browser. Now in beta as a research preview, you can assign multiple coding tasks to Claude that run on Anthropic-managed cloud infrastructure, perfect for tackling bug backlogs, routine fixes, or parallel development work. Run coding tasks in parallel Claude Code on the web lets you kick off codi

Claude can now use Skills to improve how it performs specific tasks. Skills are folders that include instructions, scripts, and resources that Claude can load when needed. Claude will only access a skill when it's relevant to the task at hand. When used, skills make Claude better at specialized tasks like working with Excel or following your organization's brand guidelines. You've already seen Ski

Claude Haiku 4.5, our latest small model, is available today to all users. What was recently at the frontier is now cheaper and faster. Five months ago, Claude Sonnet 4 was a state-of-the-art model. Today, Claude Haiku 4.5 gives you similar levels of coding performance but at one-third the cost and more than twice the speed. Claude Haiku 4.5 even surpasses Claude Sonnet 4 at certain tasks, like us

In a joint study with the UK AI Security Institute and the Alan Turing Institute, we found that as few as 250 malicious documents can produce a "backdoor" vulnerability in a large language model—regardless of model size or training data volume. Although a 13B parameter model is trained on over 20 times more training data than a 600M model, both can be backdoored by the same small number of poisone

Claude Code now supports plugins: custom collections of slash commands, agents, MCP servers, and hooks that install with a single command. Share your Claude Code setup with plugins Slash commands, agents, MCP servers, and hooks are all extension points you can use to customize your experience with Claude Code. As we've rolled them out, we've seen users build increasingly powerful setups that they

Petri: An open-source auditing tool to accelerate AI safety research Petri (Parallel Exploration Tool for Risky Interactions) is our new open-source tool that enables researchers to explore hypotheses about model behavior with ease. Petri deploys an automated agent to test a target AI system through diverse multi-turn conversations involving simulated users and tools; Petri then scores and summari

Published Sep 29, 2025 The Claude Agent SDK is a collection of tools that helps developers build powerful agents on top of Claude Code. In this article, we walk through how to get started and share our best practices. Last year, we shared lessons in building effective agents alongside our customers. Since then, we've released Claude Code, an agentic coding solution that we originally built to supp

Published Sep 29, 2025 Context is a critical but finite resource for AI agents. In this post, we explore strategies for effectively curating and managing the context that powers them. After a few years of prompt engineering being the focus of attention in applied AI, a new term has come to prominence: context engineering. Building with language models is becoming less about finding the right words

Today, we’re introducing new capabilities for managing your agents’ context on the Claude Developer Platform: context editing and the memory tool. With our latest model, Claude Sonnet 4.5, these capabilities enable developers to build AI agents capable of handling long-running tasks at higher performance and without hitting context limits or losing critical information. Context windows have limits

We’re introducing several upgrades to Claude Code: a native VS Code extension, version 2.0 of our terminal interface, and checkpoints for autonomous operation. Powered by Sonnet 4.5, Claude Code now handles longer, more complex development tasks in your terminal and IDE. Claude Code on more surfaces VS Code extension We’re introducing a native VS Code extension in beta that brings Claude Code dire

Claude Sonnet 4.5 is the best coding model in the world. It's the strongest model for building complex agents. It’s the best model at using computers. And it shows substantial gains in reasoning and math. Code is everywhere. It runs every application, spreadsheet, and software tool you use. Being able to use those tools and reason through hard problems is how modern work gets done. Claude Sonnet 4

Today, Microsoft announced that Claude models are now available in Microsoft 365 Copilot, bringing Claude to millions of enterprise users through Microsoft's productivity platform. Starting with Microsoft's Researcher agent and Copilot Studio, enterprise customers can now choose Claude Sonnet 4 and Opus 4.1 alongside existing model options, giving organizations greater flexibility to select the ri

Published Sep 17, 2025 This is a technical report on three bugs that intermittently degraded responses from Claude. Below we explain what happened, why it took time to fix, and what we're changing. Between August and early September, three infrastructure bugs intermittently degraded Claude's response quality. We've now resolved these issues and want to explain what happened. In early August, a num

Published Sep 11, 2025 Agents are only as effective as the tools we give them. We share how to write high-quality tools and evaluations, and how you can boost performance by using Claude to optimize its tools for itself. The Model Context Protocol (MCP) can empower LLM agents with potentially hundreds of tools to solve real-world tasks. But how do we make those tools maximally effective? In this p

Claude can now create and edit Excel spreadsheets, documents, PowerPoint slide decks, and PDFs directly in Claude.ai and the desktop app. This transforms how you work with Claude—instead of only receiving text responses or in-app artifacts, you can describe what you need, upload relevant data, and get ready-to-use files in return. File creation is now available as a preview for Max, Team, and Ente

Today, we're rolling out updates to our Consumer Terms and Privacy Policy that will help us deliver even more capable, useful AI models. We're now giving users the choice to allow their data to be used to improve Claude and strengthen our safeguards against harmful usage like scams and abuse. Adjusting your preferences is easy and can be done at any time. These updates apply to users on our Claude

We’ve developed sophisticated safety and security measures to prevent the misuse of our AI models. But cybercriminals and other malicious actors are actively attempting to find ways around them. Today, we’re releasing a report that details how. Our Threat Intelligence report discusses several recent examples of Claude being misused, including a large-scale extortion operation using Claude Code, a

We've spent recent months connecting Claude to your calendar, documents, and many other pieces of software. The next logical step is letting Claude work directly in your browser. We view browser-using AI as inevitable: so much work happens in browsers that giving Claude the ability to see what you're looking at, click buttons, and fill forms will make it substantially more useful. But browser-usin

Enterprise and Team customers can now upgrade to premium seats that include more usage and Claude Code—bringing our app and powerful coding agent together under one subscription. Users can move seamlessly between ideation and implementation, while admins get the visibility and controls they need to scale Claude across their organization. We are also introducing a new Compliance API, giving organiz

Claude Sonnet 4 now supports up to 1 million tokens of context on the Anthropic API—a 5x increase that lets you process entire codebases with over 75,000 lines of code or dozens of research papers in a single request. Long context support for Sonnet 4 is now in public beta on the Anthropic API and in Amazon Bedrock, with Google Cloud’s Vertex AI coming soon. Longer context, more use casesWith long