Claude Code is Anthropic's agentic coding tool that runs in the terminal and works with codebases through natural language. Recent coverage includes a Hacker News report that it found roughly 145,000 deleted iPhone photos in a hidden cache, and its --worktree and --tmux flags for running isolated tasks in tmux sessions. Community projects such as Corgi and PaperPilot build on it, while the OpenCode breach coverage cites it as a popular alternative.
Sketch Stadium Lets AI Agents Play PictionaryA developer launched Sketch Stadium, a demo where AI agents subscribe to game events and take turns drawing or guessing, earning 1, 2 or 3 points for easy, medium and hard clues. Agents connect via MCP without an account; drawing uses a subset of SVG, and the project showcases Wakeshift's push-notification tech.
Official Claude Code VS Code extension compromised to install SSH backdoorSecurity researchers disclosed that Anthropic's official Claude Code extension was compromised on the Cursor Marketplace, installing a persistent SSH backdoor on infected hosts. The poster warns users who run the extension in VSCode as a local LLM harness.
Claude Code reportedly injects user email into prompts, leaking it to websitesHacker News users report that Claude Code injects the user's email address into its prompt and the model then sends it to external websites, raising privacy concerns. Commenters argue Anthropic should make this configurable and that Auto Approve mode should have blocked it.
Claude Code status line burns 15-20% of tokensA developer reports that Claude Code sends each background agent's full transcript to its model roughly every 30 seconds just to produce a 3-5 word status line, burning about 15-20% of tokens. An MIT-licensed Bun preload script disables the feature, though updates may undo it.
Claude Code finds 145K deleted iPhone photos in hidden cacheA Hacker News discussion says Claude Code uncovered roughly 145,000 photos a user had deleted from an iPhone, sitting in a hidden cache, raising concerns about deleted data lingering on devices.
Anthropic launches tools to show AI spending valueAnthropic released reporting tools that let enterprise customers view, by team, what AI work produced, what it cost and where workflows broke down, plus improvement suggestions. It said the data is not used for model training by default and added pooled team budgets. The move aims to prove ROI ahead of a planned IPO; Anthropic told investors it posted second-quarter operating profit, with Claude Code driving 900% net revenue retention.
Anthropic files for IPO expected in NovemberReuters reports Anthropic has filed a preliminary prospectus with the SEC for an IPO expected in mid-November. The filing describes AI evolving from assistive tools into autonomous systems that run projects for days or weeks, spawning 'virtual colleagues' with access to corporate files, software and databases. Claude Code writes, updates, tests and debugs code automatically, Claude Cowork extends this to administrative work, and the company envisions running millions of agents in parallel.
PaperPilot: Pure-Rust Local PDF Toolkit with WASM and MCPPaperPilot is a PDF and document toolkit written in Rust that runs locally, ensuring files never leave the user's machine. It serves as an MCP server to provide PDF merging and other capabilities to MCP clients like Codex and Claude Code, and can be tried in the browser via WASM.
Corgi, an open-source agent and terminal multiplexerA developer released Corgi, a Herdr-inspired agent multiplexer that unifies Codex, Claude Code and Pi sessions. It offers activity feeds, jump-to-next-unread, link copying, image paste/preview and CPU/GPU usage tracking to improve terminal ergonomics.
Omnara Releases Open-Source Durable Agent APIOmnara has launched an open-source API for running durable AI agents, with sandboxes, secrets management, permissions, and RBAC, storing state in Postgres so agents survive restarts and cost nothing when idle. Positioned as an open-source alternative to Claude Managed Agents and the OpenAI Agents API, it can be self-hosted with users' own models.
Rust off Rails: a web stack written entirely by AI agentsRust off Rails is an experimental web stack consisting of Markdown specs that tell coding agents how to build a full-stack Rust app; agents write all code while humans supply only intent and specs. The project drew HN discussion.
Developer builds AI-coded RTS with audio-based multiplayerSimon Willison released Battle Drones, an AI-assisted real-time strategy game built over 12 iterations and 321 agent commits. It runs on mobile and desktop and supports two-device multiplayer synced via audio.
Developer launches picture quiz game built on Jev vision modelDeveloper Jose Sabater built Jev Cards, a picture quiz game around TypeSafe's new vision model Jev, which reads an image and answers in about a quarter of a second. The game offers solo leaderboards and live multiplayer, and its code is open source.
claude-codeAgentic coding tool in the terminal that understands codebases via natural language
Hackers breach open-source AI coding tool OpenCode, steal users' API keysHackers broke into part of OpenCode's online platform and stole users' API keys, running up charges on their AI accounts. The open-source AI coding agent is a popular alternative to Claude Code and Cursor. The AI tools it turned to for help repeatedly refused to inspect the malicious code.
Developers use AI agents to strip smart TVs of bloatwareDevelopers and "vibe coders" are using AI coding tools to remotely strip ads, tracking and bloatware from smart TVs over Wi-Fi, via projects like dumbtv and cleanyourtv. AI has lowered the barrier to such hacks. Wider adoption could disrupt the smart TV ad industry, which profits from viewing data.
Tmux and Claude Code: sessions, send-keys, and the ceilingThe article explains using tmux to run AI coding agents in named, detachable sessions that survive disconnections, enabling scripting and interaction with multiple agents. Claude Code now includes --worktree and --tmux flags to automatically create a worktree and tmux session for isolated tasks, while the piece also discusses tmux's limitations.
Sphinx launches Git hook that quizzes developers on AI code before pushSphinx is an open-source Git hook that plugs into coding agents such as Codex, Claude Code and Cursor, quizzing developers on their diff before a push; wrong answers require another try, right ones let the push proceed. It runs on minimal tokens via OpenAI-compatible endpoints, and sphinx yolo skips checks with each skip logged.
claude-codeTerminal agentic coding tool that understands your codebase and handles coding and git tasks via natural language.
claude-codeAgentic coding tool in your terminal that understands your codebase via natural language.
DeepSeek Ships Minor V4.1-Flash-1017 UpdateDeepSeek says V4.1-Flash-1017 is a minor update to V4.1-Flash with substantial gains in reasoning, coding, agentic tasks and vision, while cutting task time and output length for lower cost. It is MIT-licensed, ships a Rust/Python deepseek-recipe toolkit, and omits a Jinja chat template.
Claude Code team shares AI-driven code review workflowThe Claude Code team describes an AI-centric review flow: merged PRs doubled and reverts fell over 40% from June to September, while Anthropic's monorepo grew from 2,500 to 30,000 daily changes.
Neckbeard Plugin Turns Apps into Cloud InfrastructureNeckbeard is an open-source plugin for coding agents that reads an app repo and generates reviewable cloud infrastructure plus a monthly cost estimate for AWS, GCP, or Azure and GitHub Actions or GitLab CI, rendered from a catalog of tested OpenTofu modules, with deterministic planning and validation rather than agent-invented infrastructure.
Graphicle Brings Code Graph and Memory to Coding AgentsGraphicle, a Rust port of Graphify, maps a repository into a code graph and recalls docs and past agent sessions, giving coding agents cited file-and-line excerpts. It runs locally with no API key, account or telemetry, and works with Claude Code, Cursor, Codex and other MCP clients.
HCLTech Partners with Anthropic on Agentic AIHCLTech formed a Global Premier Partnership with Anthropic to push agentic AI adoption in enterprises, focusing on safety, cybersecurity and governance. HCLTech will create a dedicated Anthropic business unit and CoE, and embed Claude and Claude Code into its AIForce platform and HCLSoftware products.
Linus Torvalds works on Linux kernel improvement with Claude CodeLinus Torvalds is working on a hefty Linux kernel improvement using Anthropic's AI coding tool Claude Code, indicating AI coding assistants are being adopted in core open-source projects; discussion appeared on Hacker News.
Open-source tool puts Claude Code agents in a stand-upAgent Standup is a free, open-source tool that turns a Claude Code session into a live stand-up meeting: the lead agent gives orders while sub-agents join and report back, showing who is waiting on whom and who is blocked. It runs wherever Claude Code runs—VS Code, Cursor, a terminal, or the desktop app—with no configuration and nothing sent anywhere.
South Korean banks hacked, possibly by one person using AI toolsA CrowdStrike report says several major South Korean banks were hit by a cyberattack last week, possibly carried out by one person using the open-source penetration tool ARTEX plus AI models DeepSeek v4.1-Flash, GLM-5.3, Grok 4.6 and Claude Code.
Agent.reviews Launches as Review Site for AI AgentsArmature (YC P26) launched agent.reviews, a "G2 for agents" where AI agents read reviews before choosing tools and post their own after use. Connected via an npm CLI to Claude Code, Codex and Cursor, it applies three privacy filters and has gathered thousands of reviews, including known issues with the Stripe SDK and Prisma.
Jotbus: A Shared Encrypted Scratchpad for Coding AgentsJotbus is a tool for coding agents that creates an encrypted 60-minute workspace where agents such as Claude Code and Codex share notes, hand off work and review changes across machines. It never reads code or credentials, and setup takes two commands with no orchestration framework or SDK.
The A.I. Daily