AI News
Signal over noise. Daily.
一線 AI Lab
17 articlesWeek 14 · March 30 – April 3, 2026
Computer use in the CLI, interactive in-product lessons, flicker-free rendering, per-tool MCP result-size overrides, and plugin executables on PATH. Releases v2.1.86 → v2.1.91 5 features · March 30 – April 3 Computer use in the CLI research preview Last week computer use landed in the Desktop app. This week it's in the CLI: Claude can open native apps, click through UI, test its own changes, and fix what breaks, all from your terminal. Web apps already had verification loops; native iOS, macOS, and other GUI-only apps didn't. Now they do. Best for closing the loop on apps and tools where there's no API to call. Still early; expect rough edges.
Week 23 · June 1–5, 2026
Run auto mode on Amazon Bedrock, Google Cloud's Agent Platform, and Microsoft Foundry, prompt before writing files that can run code in acceptEdits mode, list installed plugins with /plugin list, and require an approved version range for managed deployments. Releases v2.1.158 → v2.1.165 4 features · June 1–5 Auto mode on Amazon Bedrock, Google Cloud's Agent Platform, and Microsoft Foundry v2.1.158 Auto mode is now available on Amazon Bedrock, Google Cloud's Agent Platform, and Microsoft Foundry for Opus 4.7 and Opus 4.8, replacing permission prompts with background safety checks on third-party providers. Opt in by setting CLAUDE\_CODE\_ENABLE\_AUTO\_MODE=1 .
Run Claude Code programmatically
Use the Agent SDK to run Claude Code programmatically from the CLI, Python, or TypeScript. The Agent SDK gives you the same tools, agent loop, and context management that power Claude Code. It's available as a CLI for scripts and CI/CD, or as Python and TypeScript packages for full programmatic control. To run Claude Code in non-interactive mode, pass `-p` with your prompt and any CLI options: claude -p "Find and fix the bug in auth.py" --allowedTools "Read,Edit,Bash" This page covers using the Agent SDK via the CLI (`claude -p`). For the Python and TypeScript SDK packages with structured outputs, tool approval callbacks, and native message objects, see the full Agent SDK documentation. Add the `-p` (or `--print`) flag to any `claude` command to run it non-interactively. All CLI options…
Manage costs effectively
Track token usage, set team spend limits, and reduce Claude Code costs with context management, model selection, extended thinking settings, and preprocessing hooks. Claude Code charges by API token consumption. For subscription plan pricing (Pro, Max, Team, Enterprise), see claude.com/pricing. Per-developer costs vary widely based on model selection, codebase size, and usage patterns such as running multiple instances or automation. Across enterprise deployments, the average cost is around \$13 per developer per active day and \$150-250 per developer per month, with costs remaining below \$30 per active day for 90% of users. To estimate spend for your own team, start with a small pilot group and use the tracking tools below to establish a baseline before wider rollout. This page covers…
Agent SDK reference - TypeScript
Complete API reference for the TypeScript Agent SDK, including all functions, types, and interfaces. npm install @anthropic-ai/claude-agent-sdk The SDK bundles a native Claude Code binary for your platform as an optional dependency such as `@anthropic-ai/claude-agent-sdk-darwin-arm64`. You don't need to install Claude Code separately. If your package manager skips optional dependencies, the SDK throws `Native CLI binary for not found`; set `pathToClaudeCodeExecutable` to a separately installed `claude` binary instead. When you compile your application into a single-file executable with `bun build --compile`, the SDK cannot resolve the bundled CLI binary at runtime. `require.resolve` does not work inside the compiled executable's `$bunfs` virtual filesystem, so the SDK throws `Native CLI…
Subagents in the SDK
Define and invoke subagents to isolate context, run tasks in parallel, and apply specialized instructions in your Claude Agent SDK applications. Subagents are separate agent instances that your main agent can spawn to handle focused subtasks. Use them to isolate context, run multiple analyses in parallel, and apply specialized instructions without adding to the main agent's prompt. This guide explains how to define and use subagents in the SDK using the `agents` parameter. You can create subagents in three ways: * **Programmatically**: use the `agents` parameter in your `query()` options. See the TypeScript and Python references * **Filesystem-based**: define agents as markdown files in `.claude/agents/` directories. See defining subagents as files * **Built-in general-purpose**: Claude…
Intercept and control agent behavior with hooks
Intercept and customize agent behavior at key execution points with hooks Hooks are callback functions that run your code in response to agent events, like a tool being called, a session starting, or execution stopping. With hooks, you can: * **Block dangerous operations** before they execute, like destructive shell commands or unauthorized file access * **Log and audit** every tool call for compliance, debugging, or analytics * **Transform inputs and outputs** to sanitize data, inject credentials, or redirect file paths * **Require human approval** for sensitive actions like database writes or API calls * **Track session lifecycle** to manage state, clean up resources, or send notifications This guide covers how hooks work and how to configure them, with examples for common patterns like…
Handle approvals and user input
Surface Claude's approval requests and clarifying questions to users, then return their decisions to the SDK. While working on a task, Claude sometimes needs to check in with users. It might need permission before deleting files, or need to ask which database to use for a new project. Your application needs to surface these requests to users so Claude can continue with their input. Claude requests user input in two situations: when it needs **permission to use a tool** (like deleting files or running commands), and when it has **clarifying questions** (via the `AskUserQuestion` tool). Both trigger your `canUseTool` callback, which pauses execution until you return a response. This is different from normal conversation turns where Claude finishes and waits for your next message. For…
How the agent loop works
Understand the message lifecycle, tool execution, context window, and architecture that power your SDK agents. The Agent SDK lets you embed Claude Code's autonomous agent loop in your own applications. The SDK is a standalone package that gives you programmatic control over tools, permissions, cost limits, and output. You don't need the Claude Code CLI installed to use it. When you start an agent, the SDK runs the same execution loop that powers Claude Code: Claude evaluates your prompt, calls tools to take action, receives the results, and repeats until the task is complete. This page explains what happens inside that loop so you can build, debug, and optimize your agents effectively. Every agent session follows the same cycle:
Agent SDK reference - Python
Complete API reference for the Python Agent SDK, including all functions, types, and classes. Install the package into a virtual environment. On recent Debian, Ubuntu, and Homebrew Python installs, running `pip install` against system Python fails with `error: externally-managed-environment`. python3 -m venv .venv source .venv/bin/activate pip install claude-agent-sdk For uv, Windows PowerShell, and API key setup, see Get started in the Agent SDK overview. The Python SDK provides two ways to interact with Claude Code: | Feature | `query()` | `ClaudeSDKClient` | | :------------------ | :--------------------------------------------- | :--------------------------------- | | **Session** | Creates a new session by default | Reuses same session |
Agent SDK overview
Build production AI agents with Claude Code as a library Build AI agents that autonomously read files, run commands, search the web, edit code, and more. The Agent SDK gives you the same tools, agent loop, and context management that power Claude Code, programmable in Python and TypeScript. import asyncio from claude_agent_sdk import query, ClaudeAgentOptions async def main(): async for message in query( prompt="Find and fix the bug in auth.py", options=ClaudeAgentOptions(allowed_tools=["Read", "Edit", "Bash"]), ): print(message) # Claude reads the file, finds the bug, edits it asyncio.run(main()) import { query } from "@anthropic-ai/claude-agent-sdk"; for await (const message of query({ prompt: "Find and fix the bug in auth.ts", options: { allowedTools: ["Read", "Edit", "Bash"] }
Connect to external tools with MCP
Configure MCP servers to extend your agent with external tools. Covers transport types, tool search for large tool sets, authentication, and error handling. The Model Context Protocol (MCP) is an open standard for connecting AI agents to external tools and data sources. With MCP, your agent can query databases, integrate with APIs like Slack and GitHub, and connect to other services without writing custom tool implementations. MCP servers can run as local processes, connect over HTTP, or execute directly within your SDK application. This page covers MCP configuration for the Agent SDK. To add MCP servers to the Claude Code CLI so they load in every project, see MCP installation scopes. This example connects to the Claude Code documentation MCP server using HTTP transport and uses…
How Claude Code works
Understand the agentic loop, built-in tools, and how Claude Code interacts with your project. Claude Code is an agentic assistant that runs in your terminal. While it excels at coding, it can help with anything you can do from the command line: writing docs, running builds, searching files, researching topics, and more. This guide covers the core architecture, built-in capabilities, and tips for working effectively. For step-by-step walkthroughs, see Common workflows. For extensibility features like skills, MCP, and hooks, see Extend Claude Code. When you give Claude a task, it works through three phases: **gather context**, **take action**, and **verify results**. These phases blend together. Claude uses tools throughout, whether searching files to understand your code, editing to make…
Verifying Rust cryptography in SymCrypt, from standards to code
How Rust, Lean, Aeneas, and AI agents are helping scale formal verification for production cryptographic algorithms At a glance SymCrypt develops new verified cryptography using Rust, Aeneas, and Lean to provide higher security assurance. We prove that their code safely and correctly implements standard algorithms, notably for post-quantum cryptography. We are releasing verified code, specs, properties, and proofs initially for SHA-3 and ML-KEM. Aeneas allows verifying a large subset of Rust code and provides efficient automation in Lean to support the proof effort. Agents allow scaling automation by writing proofs that are independently-verifiable. Introduction and motivation for formal verification Cryptographic code sits at the foundation of modern computing. It protects operating…
Claude For Creative Work
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
Theo Hourmouzis General Manager Australia New Zealand
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
Claude Design Anthropic Labs
Today, we’re launching Claude Design, a new Anthropic Labs product that lets you collaborate with Claude to create polished visual work like designs, prototypes, slides, one-pagers, and more.
媒體
22 articlesVideo generation startup PixVerse raises $439M, valuation soars past $2B
Singapore-based video generation startup PixVerse closed a Series C extension on the strength of 15 million monthly active users, it said.
Hermes agent maker Nous Research in talks for new funding at $1.5B valuation
The company is raising at least $75 million, led by Robot Ventures, with significant participation from USV and other prominent investors.
Siri AI Is Becoming Apple’s Everything Tool
Apple’s revamped Siri is more than a voice assistant; it’s now the backbone of the iPhone user experience. You can try it now through the iOS 27 public beta.
Satya Nadella has issued a shocking warning to companies using AI
Of all the debates raging about the potential downsides of AI, there is one worry causing the most hand-wringing among AI enthusiasts in Silicon Valley — that the giant AI labs that sell proprietary models are somehow acting like Trojan horses.
Siri AI is already changing how I use my iPhone
Siri AI in iOS 27. iOS 27 escaped the developer world today with the launch of the first public beta. I've been testing the new operating system since early June, looking for quirks and seeing if it can live up to the hype Apple promised in the keynote. This year's iOS upgrades are what one might call a Snow Leopard update . That means it's light on new features and instead focused on fixing things that were broken and speeding up processes across the OS. App launches, Photos search results, and AirDrop transfers should all be faster. The Messages app now supports in-line replies and end-to-end encryption for RCS messages. Liquid Glass has gotten more refined, … Read the full story at The Verge.
Labels for Music Created With A.I. Could Join Explicit Lyrics Warnings
Major music industry groups, including the organization behind the Grammy Awards, have proposed labels for tracks made with some degree of artificial intelligence.
Should You Be Polite to A.I.?
Should you say please and thank you to chatbots? This week on “Hard Fork,” Kevin and Casey talk with professor and A.I. expert Jeff Sebo about why he feels it’s important to be nice to your future robot overlords.
Apple sues OpenAI after ex-engineer allegedly used bug to steal trade secrets
Apple is gunning for OpenAI, demanding steep penalties after stumbling on a “rare” bug that temporarily allowed a poached employee that joined OpenAI to maintain access to confidential information on Apple servers for weeks after his termination. In a lawsuit filed Friday, Apple sought several injunctions blocking OpenAI from using confidential information allegedly stolen by former employees. According to Apple’s complaint, OpenAI conspired with former Apple employees as part of a grand scheme to “take an unlawful shortcut” and launch a line of AI-powered devices as marketable as Apple’s iPhone. Apple explained that it found a bug while investigating internal messages between a then-current employee, Yu-Ting “Alyssa” Peng, and an engineer who spent eight years “working on some of Apple’s…
The wildest allegations in Apple’s trade secrets lawsuit against OpenAI
Apple’s trade secrets lawsuit against OpenAI contains allegations that range from employees joking about unauthorized access to Apple’s systems to claims that job candidates were asked to bring Apple hardware to interviews. Here are the complaint’s most eye-catching claims.
Trump Administration Is Snapping Up Stakes in Private Companies. Could A.I. Be Next?
Some tech executives are nervous that the administration’s recent scrutiny of artificial intelligence models could be a prelude to a demand for an ownership stake.
What Anthropic’s latest AI discovery does—and doesn’t—show
This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here . Anthropic—currently the world’s most valuable AI company, with a nearly $1 trillion valuation—has a reputation for publishing strange and heady research. It’s looking into whether AI models can feel pain , for example, and will sometimes cut off chatbot conversations if it suspects users are “abusing” the model. One niche that Anthropic spends more time and money on than other AI companies is called mechanistic interpretability, which means looking inside the complex math of an AI model to learn why it comes up with one particular output and not another. It’s complicated stuff; there are millions of data points that might contribute to any result, and…
Sam Altman’s space data center trash talk is what most experts already believe
Responding to Musk accusing him of being a scammer, Altman said, "homeboy you're the one sellling [sic] public market investors on short-term space datacenters."
The 6 wildest claims in Apple’s lawsuit against OpenAI
When Apple employees interviewed for jobs at OpenAI, the AI startup's hardware head allegedly asked them to show up with something unusual: components they were working on and unreleased product samples. That's according to a blockbuster lawsuit filed by Apple, which accuses OpenAI of stealing confidential documents , spying on hardware prototypes, and tricking one of its trusted partners into performing a proprietary product design technique. The lawsuit primarily revolves around the alleged actions of three people: Tang Tan, a 24-year Apple veteran who recently served as the vice president of the Apple Watch. In 2024, Tan left to work on … Read the full story at The Verge.
Should AI help you get away with killing your spouse?
What does a world of total user-aligned AI actually look like?
Anthropic starts localizing Claude pricing for India, its biggest market after the US
Claude users in India are starting to see Indian rupee-denominated subscription plans.
Now, defenders are embracing the prompt injection, too
Prompt injections, the malicious commands attackers embed into content to entice large language models to follow them, have been attackers’ go-to tool for turning AI platforms against their users. A well-phrased command sneaked into an email or calendar invitation is often all it takes to cause the LLM to exfiltrate sensitive data or follow other harmful actions. Now, defenders are embracing the prompt injection, too. A strong, sharp effect Researchers from Tracebit on Monday said they found that placing prompt injections alongside passwords, cryptographic keys, and other secrets stored on Amazon Web Services was often all that was needed to shut down attacks from AI hacking agents. The prompts direct the attacking LLM to perform an action forbidden by its guardrails, the safety barriers…
What the Fight With Anthropic Reveals About Free Speech in America
We won’t get the A.I. regulation we deserve until we change how we think about the right to code.
Waze adds new AI-powered features and customization updates
Some of the new features are powered by Google's Gemini AI assistant, which reflects the tech giant's broader push to integrate Gemini across its products while also better positioning Waze to compete with rival services such as Apple Maps.
Nearly 200 Economists and Tech Leaders Warn of A.I. Threats
A letter calls for policymakers to do more to understand and respond to potential disruptions from artificial intelligence.
Wall Street’s Big Week for Earnings and Economic Data
Earnings season kicks off in earnest on Tuesday with big banks up first. Corporate America faces a high bar to beat investor expectations.
Simulating everything, sort of: The promise and limits of world models
Over the past few years, many of us have gotten a crash course in what we now call artificial intelligence—but really, it has mostly been a crash course in large language models. Increasingly, however, LLMs are no longer the only category of AI drawing high expectations, massive funding rounds, and significant research and product development. Over the past year, we've seen a plethora of new announcements in a category labeled "world models," and you'll likely see more movement there in the coming months and years. Instead of or in addition to working with language, world models aim to lay the groundwork for AI systems that are capable of simulating the physical world, or at least a useful approximation of it. Read full article Comments
Waze is getting a bunch of new AI-powered features
Waze is getting an AI makeover. Google is integrating its flagship AI assistant, Gemini, into the driving app with the goal of letting users personalize their trips a little more. Of the four new updates, only two are being described as involving Gemini. Waze says its updating its conversation reporting feature, first introduced in 2024 , to allow drivers to use conversational voice commands to report traffic incidents and suggest map updates, like a road closure or outdated house number. In addition, Waze introduced Destination Search, enabling drivers to use (again) use conversation voice commands (e.g., "Find me a coffee shop that's open … Read the full story at The Verge.