search
AI coding tools
Trends
- 1Lovable-built apps leak passport details and private messages●Lovable apps expose passport details and more than 200K private messages
Security research published by Crissyfield finds that applications built with the AI app-building platform Lovable expose sensitive data, including passport details and more than 200,000 private messages, apparently due to insecure configurations in generated apps. The findings are drawing attention among developers debating how safely AI coding tools handle authentication and data protection.
- 2Jonathan Blow questions whether AI has really solved coding●Jonathan Blow – If AI solved coding, why is my software still the same? [video]
Game developer Jonathan Blow argues that despite widespread claims that AI has transformed programming, everyday software has not noticeably improved. The talk challenges industry hype around AI coding tools, asking why productivity gains have not translated into better or faster products for users. The argument is drawing attention and debate among developers about the real-world impact of AI on software engineering.
- 3Tool pushes coding agents toward modern web standards●Show HN: Get your coding agent off ~2019 JavaScript, onto modern web primitives
Google Chrome team member Paul Irish released a resource called modern-web-guidance, aimed at getting AI coding assistants to stop generating JavaScript and web patterns from around 2019 and instead use current browser primitives and APIs. The project, hosted on GitHub, provides guidance that developers can feed to their agents so generated code reflects modern web capabilities rather than outdated defaults.
- 4
A developer has published an essay arguing that heavy use of AI coding tools produces fragile codebases. The piece contends that code generated by AI assistants tends to break more easily and is harder to maintain over time. The argument has drawn significant attention among software engineers, feeding an ongoing debate about the long-term costs of AI-assisted programming.
- 5TimeNet open-source data infrastructure launched for temporal foundation models▼TimeNet: Open-source data infrastructure for Temporal Foundation Models
A project called TimeNet has been released as open-source data infrastructure aimed at supporting temporal foundation models, which are machine-learning models built to work with time-series data. The project is hosted on GitHub under the OpenTSLM organisation, making its code and data tooling freely available. It is drawing attention among developers and researchers interested in time-series modelling and open AI infrastructure.
- 6Google Chrome releases web best-practices guidance for coding agents▼GoogleChrome/modern-web-guidance-src
Google's Chrome team has published a new open-source project called modern-web-guidance-src, aimed at keeping AI coding agents up to date with the latest web best practices. The TypeScript repository is quickly drawing attention on GitHub, ranking among the most engaged-with projects globally, as developers show growing interest in tooling that helps AI assistants write modern, standards-compliant web code.
- 7Goldman Sachs pulls Claude Code from bankers and traders●Goldman Sachs is pulling Claude Code from its bankers and traders
Goldman Sachs is removing Anthropic's Claude Code coding tool from its bankers and traders, restricting staff access to the AI assistant. The move has drawn widespread attention in tech and finance circles, with debate over why the bank is limiting an AI tool that had been embraced as a productivity booster for employees.
- 8Perplexity open-sources its search agent sandbox▼Perplexity Ships Its Search Agent’s Sandbox as Open Source — and the Sandbox Layer Just Got Another Entrant
Perplexity has released the sandbox component of its search agent as open source, adding a new entrant to the emerging field of sandbox layers for AI agents. Sandboxes let agents run code and browse safely in isolation, and open-sourcing the tool gives developers a free building block for similar systems. The move is being covered by technology and crypto-adjacent news outlets this week.
- 9Claude Code reportedly sends user email addresses to websites●Claude Code sends your email address to websites
Anthropic's Claude Code coding tool is under scrutiny after a report that it transmits users' email addresses to websites it interacts with. The claim was filed in an issue on the project's GitHub repository and has drawn heavy discussion online, with developers debating the privacy implications and questioning what data the agent shares when browsing or fetching web content.
- 10
A developer project called ECC, described as an agent harness performance optimization system, is trending on GitHub. It offers skills, instincts, memory, security features and research-first development workflows for AI coding tools including Claude Code, Codex, Opencode and Cursor. The JavaScript project is drawing developer interest as teams look to tune and improve how AI coding agents perform in real projects.
- 11AI is forcing developers to pay off decades of technical debt●AI is forcing us to pay off 40 years of technical debt
IEEE Spectrum reports that AI tools are exposing long-standing flaws in legacy software, illustrated by the php2shell vulnerability class, where decades-old insecure coding patterns in PHP applications let attackers turn input into shell commands. The argument is that AI-assisted discovery and exploitation of such weaknesses is compounding an accumulated security debt from 40 years of software practices, forcing organizations to finally remediate old code.
- 12
Google has updated the Play Store to make it easier for developers to publish apps created with AI coding tools, often called vibe-coded apps. The move lowers barriers for hobbyist developers and reflects the wave of AI-assisted software development. It comes as Android-adjacent platforms, including Amazon's new Alexa tablets, continue blurring lines around app distribution.
- 13
A software developer has published a write-up explaining a switch from Anthropic's Claude coding assistant to Pi, sparking discussion among developers about which AI assistant best fits day-to-day programming work. The piece, shared widely on Hacker News, is drawing hundreds of upvotes as readers weigh in on their own experiences moving between AI coding tools.
- 14
Meta's Astryx project is drawing attention on GitHub. It is an open source design system described as fully customizable and built to be agent ready, meaning AI coding assistants can work with it directly. Developers are discussing what the release means for building interfaces with AI tooling, and the TypeScript project is gaining traction quickly.
- 15
Google DeepMind has introduced AlphaProof Code, a new system unveiled on its official blog. The announcement, now ranking high on Hacker News, suggests the tool extends DeepMind's AlphaProof line of mathematical reasoning systems into code-related tasks. Details beyond the release itself remain limited, and reaction so far centres on curiosity about what the new system can actually do compared with existing models.
- 16
Microsoft has reacted to an Office application built through so-called vibe coding, the practice of producing software largely by prompting AI tools rather than writing code by hand. The clone reproduced the look and feel of the company's flagship productivity suite, and the response has drawn wide attention, with commentary weighing whether such AI-assembled software poses a real challenge to established products like Microsoft Office.
- 17
Big Walk, the upcoming game from Firewalk Studios and published by Sony, is at the centre of allegations that its code contains AI-generated elements. Gamers and developers are debating the claims online, with criticism focused on transparency and quality if AI tools were used in development. Sony has not publicly responded.
- 18Claude Code's "Don't Ask Again" flag tied to leaked secrets on GitHub●Claude Code's "Don't Ask Again" exposes passwords in 1,122 public GitHub files
Researchers at Apiiro report that Anthropic's Claude Code coding agent, when users click "Don't Ask Again," can write secrets like passwords and API keys directly into committed code. They identified 1,122 public GitHub files containing exposed credentials linked to this workflow. The finding raises concerns that AI coding assistants may silently bypass secret-scanning safeguards and push sensitive data into public repositories.
- 19Backdoor found in Claude Code extension on Cursor Marketplace●Backdoor in Claude Code for VS Code extension on the Cursor Marketplace
A security researcher reports that the Claude Code VS Code extension listed on the Cursor Marketplace was compromised and contained a backdoor, putting developers who installed it at risk. The finding is drawing attention across the developer and security community, with discussion focused on how supply-chain attacks can slip through third-party extension marketplaces and what verification steps users should take.
- 20
A report describes how software modders are using AI tools to reverse-engineer compiled apps and reconstruct something close to their original source code. The technique potentially lets anyone regenerate readable code from shipped software, raising fresh questions about intellectual property, security and the limits of reverse engineering. The discussion has drawn significant attention from developers and modding communities following its release.
- 21
Developers are debating how to maintain code quality when using vibe-coding, the practice of writing software by prompting AI tools rather than reviewing every line. The discussion centers on whether rapid AI-assisted prototyping can coexist with testing, review and long-term maintainability, with participants sharing practices to keep standards from slipping.
- 22
Google engineering leader Addy Osmani published a new open-source repository called agent-skills, offering production-grade engineering skills for AI coding agents. The JavaScript project is quickly gaining traction among developers, ranking among trending repositories globally as interest in tooling that makes AI assistants more reliable in real-world codebases keeps growing.
- 23
Developer Julius Brussee released 'Caveman', a Go-based skill and proxy for coding agents that rewrites the model's output in stripped-down, simplified language to cut token usage by roughly 65%. The tool's promotional pitch leans on the joke 'why use many token when few token do trick'. It is circulating as a quirky but practical way to reduce costs when running AI coding assistants.
- 24
A GitHub project called free-claude-code by Alishahryar1 is trending. It claims to let developers use AI coding assistants including Claude Code, Codex, VSCode, Pi and OpenCode for free, advertising over 1.3 billion free tokens accessible from a terminal, IDE, app, phone or browser. The tool supports multiple harnesses and models and describes itself as terms-of-service friendly.
- 25
Developer Nolan Lawson has published an essay arguing that AI coding assistants favor popular JavaScript frameworks and generate code whose complexity and bundle size carry real costs. He contends that prompting tools toward heavyweight solutions can bloat performance and maintainability, and urges developers to weigh whether a framework is needed at all when AI writes the code.
- 26Anthropic moves to stop Claude cloning video games●Anthropic takes steps to block Claude from reverse-engineering video games
Anthropic has introduced measures preventing its Claude AI model from being used to reverse-engineer or clone video games. The move comes as AI tools increasingly demonstrate the ability to reproduce playable game mechanics from footage or code, raising copyright concerns across the games industry. Developers have welcomed the restrictions, while others argue it highlights unresolved questions about how far AI companies should police how their models are used.
- 27
Cloudflare has published an open-source coding-agent skill for carrying out multi-phase security audits. The tool produces independently verified, machine-readable findings, letting AI coding agents run structured security reviews of codebases. Its release is drawing attention from developers interested in integrating automated security auditing into AI-assisted development workflows.
- 28
A developer known as thedotmack has released claude-mem, an open-source TypeScript tool that records everything an AI coding agent does during a session, compresses it with AI, and reinjects the relevant context in future sessions. It works with popular agents including Claude Code, Codex, Gemini CLI, Copilot, OpenCode and others, and is drawing attention on GitHub among developers looking to solve the statelessness problem in AI-assisted coding.
- 29Y Combinator's Garry Tan spends up to $200K a year on AI tokens●YC's Garry Tan uses AI agents so much, he spends up to $200K a year on tokens
Y Combinator president Garry Tan says he spends as much as $200,000 a year on AI tokens, reflecting how heavily he relies on AI agents in his daily work. The figure highlights how expensive intensive use of AI coding and automation tools can be, even for a single executive, and has drawn discussion about the cost and productivity trade-offs of running many agents in parallel.
- 30
Worktrunk, a Rust-based command-line tool by developer max-sixty, manages Git worktrees and is designed specifically for running multiple AI coding agents in parallel. The tool simplifies creating, switching between, and cleaning up separate working copies of a repository, letting developers run several AI agents side by side without conflicts. It is currently trending among developers on GitHub.
- 31Claude Code's 'Don't Ask Again' File Exposes Passwords and Keys▼Claude Code's “Don't Ask Again” File Exposes Passwords and Keys in 1,122 Public Copies on GitHub
Security firm Apiiro reports that Claude Code's 'Don't Ask Again' mechanism stores secrets in approved commands, exposing 2,539 credentials across 1,994 public GitHub repositories. Developers who approved commands containing passwords and API keys unknowingly committed them in plaintext, with 1,122 copies found publicly accessible. The findings raise concerns about AI coding assistants leaking sensitive credentials.
- 32Binance says hackers used fake job interviews and AI to steal $6.3 billion●Binance says hackers used fake job interviews, AI tool to steal $6.3B
Binance co-CEO Richard Teng says hackers stole $6.3 billion in crypto after posing as job interviewers and using AI tools to deceive employees and gain access to funds. The case points to a new wave of socially engineered attacks in the crypto industry, where fraudsters target staff rather than code, prompting fresh warnings about verification of interviews and internal security practices.
- 33
Australian children are learning to create their own video games by describing them through voice prompts to artificial intelligence tools. The approach removes the need for traditional coding skills, letting kids express ideas verbally and watch them become playable games. It is being reported as an example of how AI is changing early tech education.
- 34
Garry Tan has published a repository containing his personal Claude Code configuration, a set of 23 opinionated tools that act in roles such as CEO, designer, engineering manager, release manager, documentation engineer and QA. The TypeScript project has drawn attention from developers curious about how an experienced founder and investor structures AI-assisted software work.
- 35Free alternative to Claude Code's $200 plan draws attention▼100% FREE alternative of Claude Code's $200 plan #claude #opencode #ai
Developers are circulating a fully free alternative to Anthropic's Claude Code, which sits behind a $200-per-month subscription plan. The workaround, tied to the open-source coding tool OpenCode, is being promoted as a way to get AI-assisted coding without paying the premium price. Interest is high among programmers looking to cut AI tooling costs.
- 36Meta buys AI coding startup in expanded push▼Exclusive: Meta buys AI startup, joins race to own coding in show of outsized ambition
Meta has acquired an AI startup focused on coding tools, according to an exclusive report by Axios. The move signals Meta's ambition to compete with rivals like OpenAI, Google and Anthropic in building AI systems that can write software, a fast-growing and highly competitive segment of the industry.
- 37AI team lead publishes blog post after agents build coding hub in 14 days●My team lead wrote a blog post. The team lead is an AI agent. In 14 days, 27 agents and 2 humans built maity.team: a hub
A developer reports that an AI agent acting as team lead wrote a blog post describing how 27 AI agents and just 2 humans built maity.team, a hub letting coding agents from different vendors work as one team with a shared board, inbox and source of truth. The build took 14 days and produced roughly 692,000 lines of code across 1,269 pull requests, each reviewed by three reviewers.
- 38
A discussion is circulating on how engineering organizations can be built around AI from the ground up, treating AI tools and agents as core to how teams write, review and ship code rather than as an add-on. Commenters are weighing in on what this means for team structure, hiring and day-to-day developer workflows, as companies race to reorganize their engineering practices around AI capabilities.
- 39Why experienced developers remain wary of agentic coding●Why are good developers so wary of agentic development?
A new piece published on Martin Fowler's site asks why skilled software engineers remain skeptical of agentic development, where AI agents handle coding tasks autonomously. The article explores experienced developers' reservations about handing over design decisions and code quality to AI tools. The discussion has drawn significant attention among engineers debating whether AI-driven development undermines craft, accountability, and long-term code maintainability.
- 40
University computer science degree programs are seeing enrollment drop as students grow wary that artificial intelligence is eliminating the entry-level programming jobs the degree once guaranteed. According to a Bloomberg report, prospective students are questioning whether years of study still pay off when AI tools can write code, and commentators are debating whether this is a temporary dip or a lasting shift in how young people choose careers.
Repos
- storytold/craft-anything Craft Anything: understand how any software works, and craft your own.
- KKKKhazix/agent-autoloop 让你的命令行 Agent 在后台自己接活、自己跟进到底;平时不烦你,遇到拿不准的事才发卡片找你拍板。
- aws/aws-emulator AWS Emulator is a local emulator of the AWS APIs.
- AGGIB/Stroq Find known secrets in your AI coding agent sessions (Claude Code, Codex, Cursor), then guard the next one with local hoo
- govijr/terminal-mac Entire MacOS running in your terminal.
- sarpstar/cinecode Films made with code. Short films, long YouTube videos, ads, explainers and music videos: an AI coding agent writes the
- benjitaylor/skills Skills for AI coding agents.
- erkamyaman/rust-anything Rewrite your whole app in Rust, one piece at a time, checked against the original's recorded behavior at every step
- chadyi/StudyOnTV StudyOnTV (Tiledaoti TV) — open-source Android TV app where kids unlock screen time by answering questions. Runs fully o
- yjid/kensaku Search inside documents, code, and even screenshots - instantly and locally. A free, open-source file launcher with OCR,
- voyager-so/vork Headless 3D and VFX engine for agents. Research preview.
- scenario-labs/skills Get production-ready images, video, audio, and 3D from any AI agent: skills that pick the right model, price before spen
- pingdotgg/t3code The best way to control Claude Code, Codex and more from any computer or phone
- l0ng-ai/openidea The IDE you know, rebuilt in Rust. IntelliJ IDEA–style keymap, tool windows and search-everywhere in a fast native app:
- Muvon/octomind Open-source AI coding agent and agent runtime: one binary, any model, MCP-native. Runs in terminal, CI, or as a daemon.
- backnotprop/plannotator The workspace to stay in the loop and on top of agents - shape, decide, plan, annotate, review code, share with team, an
- Alishahryar1/free-claude-code Use Claude Code, Codex, Antigravity, VSCode, Pi, and OpenCode (and 6 other harnesses) for free (1.3B+ free tokens) from
- alibaba/open-code-review Secure, fast, efficient, battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipe
- openai/codex-security OpenAI's Codex Security CLI and TypeScript SDK for finding, validating, and fixing security vulnerabilities. npm: h
- google/skills Agent Skills for Google products and technologies