I have been researching what the ultimate diff tool would look like, based on actual diffs in open source repositories. OmniDiff can diff: code (25 languages), config, images, PDFs, archives, cursors, fonts... It's fast. It can diff C++ files of tens of thousands of lines in under 100ms, on a 2023 ThinkPad T14s. It's open source. AGPLv3. It's based on real diffs. I looked at the 7400 repositories that are listed in Gentoo packages, sampled 2000 diffs and solved them manually to get a good understanding of what is needed, before putting together the diff pipelines for each content type. Incidentally, this also forms the basis of the CI pipeline, so the robustness of the codebase is quite high. I'm still developing it. It's v0.3.0 at this time. But I have been using it daily as my git difftool for months now, and it works well. AI Disclosure: Yes, I do use AI. No, I do not vibecode.
ElmTUI is a declarative TUI framework for Common Lisp. I built it for fun, it has css-like styling and follows the model-update-view idea. An enhanced Lisp Repl is implemented as an example. I added capability to preserve scrollback for instances running in tmux, allowing to analyze terminal styles, or manipulate/restore earlier scrollback contents. For now its SBCL only though.
An open-source alternative to commercial Git clients, built for teams working with GitLab and GitHub: no proprietary cloud, no vendor account, no telemetry and no AI.
Look up words in the built-in macOS dictionaries, save them to your own vocabulary notebook, and review a little every day with spaced repetition. Everything stays on your Mac: no account, no ads, free. I first built it in 2013 and rebuilt v2.0 from scratch with vibe coding (pair-programming with AI).
MΛNIX Social is a free Android app to share photo/video posts, reels and 24-hour stories, chat 1:1 with friends, and shoot with an in-app camera with 7 filters. It's about 5 MB and built to run on budget phones and slow connections (Android 7+). Free with one small banner ad. Made in India by solo developer Manish Saini.
Sara combines a customizable AI companion and editable persistent memory with coding workspaces, Git and tests. Run it on your own server with local or cloud models. MIT licensed, written in TypeScript, with zero runtime dependencies.
Capture a thought and turn it into a task on iPhone. Swipe between memo cards, set date-only or timed deadlines, and review tasks in a monthly calendar or Lock Screen widgets. Notes stay on your device, with no account, ads or developer cloud sync. Includes 16 backgrounds and Japanese, English, Chinese and Spanish. Pro adds week/today timelines and photo backgrounds, with subscription or one-time purchase options.
B300, B200, and H200 are on the live list. The GPU rental price is the ask on the card. Deposit USDT, USDC, SOL, or ETH and pay for the seconds the pod runs. 亮点 B300, B200, and H200 are on the live list. The GPU rental price is the ask on the card. Deposit USDT, USDC, SOL, or ETH and pay for the seconds the pod runs.
Project URL Category JS Project Title DiffSeek: cut the noise in big GitHub PRs Project A Chrome extension for reviewing big GitHub pull requests. It filters changed files by CODEOWNERS owner, group, format and status, hides generated and already-viewed files, and shows viewed progress on every file and folder in GitHub's file tree. Highlights I'm the developer of DiffSeek. Builds on GitHub's own file tree instead of replacing it: each file and folder gets a viewed badge, and clicking a folder's badge marks its unread files viewed. Filters by every CODEOWNERS owner, by groups (built-in Tests, Docs and Config, or custom ones from .gitattributes), by format with aliases such as ts/tsx, and by viewed, generated and change status, with a live count on each row. Files marked linguist-generated in .gitattributes start hidden and are viewed automatically. No token, account or backend: it uses the existing GitHub session, so private repos work, and it collects no data. MIT-licensed, written in TypeScript. Example Code gitattributes dist/ linguist-generated=true db/migrations/ group=Migrations .tsx linguist-language=TypeScript or Videos
Othismo is a runtime which has a bunch of sandboxed WASM modules talking to each other + the host environment. It's actually hosting its own website right now at othismo.com. This is a project that's been percolating in my head for a while, as a vague desire to make a "better" runtime for software development. But it was never particularly clear for me what that would _actually_ look like. In the last month or so, I felt a bit inspired to ask what such a runtime might look like for agentic development. My hypothesis is that we really can't review 10k+ SLoC pull requests consistently and really understand them. But if we try to have a higher level view of things, we still can't be confident the agent(s) haven't done something stupid and opened a socket or a file somewhere in that 10k+ SLoC. Therefore we need a very clear demarcation between computation and side effects. In Othismo that takes the shape of the image definition files (.oth) which describe the namespace + various type definitions describing what the interface of an object (should) be. It's somewhat like TypeScript, we don't actually do anything to verify an object conforms to this interface. I'd love your feedback on the syntax used to define images overall, I feel like it doesn't do a great job giving me a clear notion of the environment lexically. I'm more comfortable about the syntax for describing the behavior of objects, although things like `operator` and `readonly` are really only necessary because I've not done anything to support auth at the protocol layer yet. You can run the same website hosted at othismo.com with `git clone [email protected]:mamidon/othismo.git` `task run:example:static` You can also play with the namespace interactively with ``` task build:example:static target/debug/othismo shell examples/static/static.rti # print out the namespace(s) target/debug/othismo --check examples/static/static.rti ``` You'll need rust, task (it's like make), and node+npm.
Build websites without a laptop. PocketCoded Web brings HTML, CSS and JavaScript editing, local preview, a terminal and Git to Android. Code and preview offline, with optional online AI assistance when you need it.
Afroai Group is a cloud IDE running on Cloudflare's edge network. Write, run, and deploy Python, JavaScript, TypeScript, Go, and Rust in your browser. Features instant startup, live collaboration, one-click deploy to Cloudflare Workers, Pages, and D1. Also includes AI website builder, mobile app builder, USSD builder, and mobile money payments (M-Pesa, MTN MoMo, Airtel Money).
Your Ultimate Place to Learn Data Analysis and AI Coding Empower your decision-making with high-fidelity insights and professional-grade development. Our
Starting a freelance WordPress career can be challenging without a clear roadmap. This guide helps beginners and aspiring developers understand the skills they need, build a professional portfolio, find clients, set competitive rates, and offer website maintenance services. It provides practical steps for turning WordPress development skills into freelance opportunities and building a sustainable business.
Free online tools for everyday work: image converters, unit converters, business, home, energy and fitness calculators, and developer tools. No sign-up.
WEBXWHALE is building modern websites, digital products, and learning tools. Explore SQLwhale to learn SQL through hands-on practice, interactive challenges, and skill assessments. Improve your coding skills, solve real-world problems, and earn a certificate with Premium access. Our mission is to make technology learning practical, accessible, and engaging for everyone. Learn, Build, and Explore with WEBXWHALE!
Zokie is a powerful agent IDE for builders and teams. Plan tasks on Jira-like boards, collaborate with AI agents and teammates, and deploy in one click.
Self-hosted remote control for Claude Code and Codex, built for teams. Everyone joins the same session from any device: one context, one environment, one task. Approvals arrive as push notifications, so whoever's online keeps the agent working and your plan's tokens don't sit idle. Chats, terminals, files, git and dev-server previews for every machine you own. Open source.
Five command-line engines for web security testing: subdomains, ports, content discovery, fingerprinting and nuclei-compatible templates. One install on Linux, macOS or Windows, a shared terminal dashboard, and JSON that pipes into anything. Apache 2.0.
I have been dabbling with programming ever since Covid was a thing. And since then I jumped around and programmed a lot of stuff, whether it was Websites, Apps or simple Python programs. Unfortunately, I have never really gotten that "feel", the feel of being useful or something like that. Most of the projects I have done, died out, and the issue was not just that my programs were not reaching anybody, it was also that I felt like I was not learning anything. Then came C, and I just have to say, that it was probably one of the best decisions I have ever made. I have tried C a lot, once I was 15, 17, and now, 19 years old. This time it stuck with me, and moved me to create my own organization called cherries.works, with the sole focus of creating efficient, open source, free and simple tools for developers. After more than 5 contributions, issues created, bugs fixed, letting it run on my own homelab for 4-5 hours, I can confidently say, that with the help of C, my biggest project Pulse was made ( https://github.com/cherries-works/pulse ). I was vacating somewhere and left my laptop at home. I brainstormed a lot about creating this organization during my vacation, and my project Pulse. This is around 6 months ago. And now after 6 months, seeing how well everything turned, really makes me happy. Might be far-fetched, but as someone who has learned multiple languages, coded in multiple areas, and created (way) too many projects, I can confidently say, that thanks to C, I have a project that actually makes me happy. Pulse is a simple monitoring tool that does not require you to sign in, or setup anything. You just download it with a single curl command, and run it on your device. I created it because I really dont like the setup hell, or whatever it takes to get a single monitoring tool running, and also because I wanted to learn C, not to mention having a program I control would mean I can add my features I would want (such as quick web hosting, API sharing, JSON formatting, styling etc.). At the end, it really was a win-win for me. Would mean a lot if devs who are interested could check it out, and point me towards some issues that I could fix, as I feel like the project is reaching a stagnation point. But even then, I am just really happy to look at Pulse, and be able to say, that *I, me* are the one who programmed that.
ASO for the App Store and Google Play, run by your agent. Ranks in 175 countries, listings in every language, store art and reviews. Pay once, on your Mac.
I built Aria because I kept making products I was proud of, but never had enough time to market them properly. Aria learns who you are, how you write, and what performs, then creates, schedules, and publishes content for you. It learns from your profile, posts, edits, and results, without copying bad writing habits. The very first Aria post crossed 100K impressions. Long-term, I want Aria to do for marketing what coding agents are doing for software.
Break free from Claude Code context limit errors and maintain peak AI dev speed. Agentic tools drain tokens fast via auto-searches. This toolkit provides a Relay Architecture to eliminate context exhaustion. What’s Inside: State Framework: CLAUDE.md &claudignore templates Handoff Prompts: Auto-compress history in seconds CLI Automation: One-line terminal aliases for instant resets Refactoring Guides: Blueprints for multi-session builds
Hi HN, I kept wiring Ben Eater's 8-bit computer wrong, so I built a realistic virtual breadboard editor. Some of the nice features are side-by-side diff, scope, schematics auto-sync, auto-router (BBRouter stands for breadboard router), and my favorites: upload from the real Arduino IDE and connect to your AI with MCP, let Claude & co build it or help you troubleshoot. The link is the SAP-1 running. You can orbit it, probe a hole with the scope, slow the clock down, play with the virtual displays for debugging, then you can actually build it following the step-by-step instructions. A side-goal is to create a community, there are quite a lot of simulators out there but mostly offline / too abstract to realistically use it for a real build. I also never found one that had a decent UI/UX and worked well on mobile or with a trackpad. Happy to answer anything, and tell me what breaks. p.s. Maker plan is $7/month; code HACKERNEWS makes the first month free and then $5/month for as long as you keep it.
Check CSV imports locally with reusable JSON rules and HTML/JSON reports. Python 3.10+ and terminal required; no packages or API. Flags missing fields, duplicates and invalid values without changing the source. $19 download. Built with AI assistance.
ASOSurge tracks your App Store and Google Play keyword ranks every day in 120+ countries, shows the keywords rivals win that you don't, and has Surge AI write a ranked weekly plan from your own data. Free plan, no developer access needed.
Your Claude plan forgets. Your Codex plan forgets. And neither has met the other. Oplyr is one window for the whole workday. A local brain that never compacts, filled as you work and handed to whichever agent you ask. Mention two in a message and they answer in turn, each reading the other. Every edit arrives as a diff you approve or send back. Beside it: a live codebase map, a real terminal, your calendar, your playlists, voice. Your code never leaves your Mac.
DevVault is an open-source, cross-platform desktop workspace bringing essential developer tools together. Manage projects, work with Git, interact with databases, test APIs, and organize your workflow in one place.
I started looking into OpenProject as an open-source JIRA alternative for personal project, and wanted a way to have my AI agents pick up tasks and complete them automatically. So I put together this OpenProject Agent Engine. It auto assigns the 'work package' (ticket), then kicks off the agent to do the work. Once complete, the agent opens a PR then leaves a detailed comment on the ticket. The engine then assigns the ticket to a QA code review agent. If it passes, the PR gets merged and and ticket is closed. If it doesn't pass, it gets kicked back to the original agent to fix the issues. Feedback welcome and hope it can help someone. Docker Hub: https://hub.docker.com/r/kookster310/op-agent-engine Github: https://github.com/310networks/openproject-agent-engine
Agentmash is an open-source multiplayer AI for teams' coding agents, a shared context layer that tells every agent which files a teammate's agent is changing, so the agents talk to each other and work on their own without overwriting anyone's code. You can switch between tasks running on different models, @mention a teammate, your own agent or a teammate's agent like in Slack, and keep skills, chats and automations in sync. Connect Claude Code, Codex, Antigravity or your own open-source models
Disclosure: I am submitting on behalf of imread. Tool: imread extracts the main text of a public web article so it can be read in a simpler interface or copied as Markdown into research notes. Core extraction works without an account. Its documented REST API and MCP endpoint also support developer workflows. A concrete example is collecting a long public article for a reading notebook while keeping the original URL beside the extracted text. The accompanying tutorial includes a small Python standard-library client that saves this attribution, refuses to overwrite an existing file, and reports extraction errors: The published example was downloaded and run against the live API before this suggestion. API documentation: Limits: login-only pages and pages dependent on client-side scripts may not extract. Speech and translation are optional interface features and may use external providers. The tutorial does not republish any third-party article body, and extraction is not a licence to redistribute a source article. This is a resource suggestion for your software/tools section, not a user review or an endorsement claim. Thank you for considering it.
iKanban is task and project management for engineering teams that work with AI coding agents. Connect Claude Code, Gemini CLI, Cursor or any MCP-compatible agent through the native MCP server, and the agents read and update the board. Sprints, releases and linked docs sit alongside the work. A security scorecard tracks OWASP Top 10 coverage for your team. Bring your own Claude, OpenAI or Gemini key. From $8 per user per month.
Claude Code and Codex in a terminal that drops down from your MacBook's notch. Live status, approvals from the notch, an office of your agents, a browser for your local server, on-device dictation and checkpoints.
Learn and build Java without a laptop. PocketCoded Java brings offline coding and on-device execution to Android, with code completion, project management, a terminal, Git and optional AI assistance.
Codex Skills is a free directory of 872 MIT-licensed Agent Skills for OpenAI Codex. Each skill is a complete SKILL.md playbook that Codex loads the moment a task needs it: PDFs, React, Postgres, security reviews, AGENTS.md and more across 28 categories. Install any skill with one command, grab curated collections like the Starter Pack, and read every line before it runs. No account, no telemetry. Works in the Codex CLI, IDE and app.
GitHub Copilot Fusion takes the guesswork out of choosing AI models for coding. Instead of picking one model, Fusion automatically routes each task to suitable models and combines them when needed. One model can write code while another reviews or improves it. Choose Standard, High, or Max effort to balance speed, quality, and cost. Available across GitHub Copilot plans in VS Code, Copilot CLI, the Copilot app, and JetBrains IDEs.
**B4Coder — Think Before You Code** B4Coder is an interactive programming education platform that helps students and beginners understand coding logic before writing code. Build programs using visual flowcharts, execute them step by step, and explore equivalent code in Python, Java, C++, C#, and JavaScript. Designed for students, beginners, and educators. **Learn visually. Understand logically. Code confidently.** Start learning at
r3 helps you review coding agent artifacts: HTML pages, Markdown docs, and code diffs. See what's built, comment on any element, passage, or line, and send it back to the agent to steer the next revision. Runs locally, open source, works with any agent.
Tapaya turns the Android phone a merchant already owns into a certified card terminal. No box to rent, no hardware to wait for: sign up online and start taking contactless payments in days. With the Accept SDK, platforms can build card acceptance straight into their own apps. Behind the simplicity sits PCI MPoC-certified technology trusted by acquirers like Nets Baltics. ➡️ Because taking a payment should feel as easy as flicking a light switch.
telemetry.dev shows each run of your AI agent as one trace. Each model call and tool call is a span with its duration, tokens, cost, and error. Prompts and evals use the same traces: change a prompt without a deploy, then run it on a dataset and compare the scores with production. Alerts tell you about errors, cost spikes, and agent loops. Connect with our TypeScript or Python SDK, or with the OpenTelemetry exporter that you already use.
Terminay keeps the shells, files and agents for each project together in one desktop app, with split layouts and coloured project tabs. Its MCP server lets Claude Code, Codex and other agent CLIs read terminal output, open splits and run commands, with permissions you set. Pair a browser or another machine for remote access. It also has cron and event automations, macros, terminal recordings and a file browser with Git diffs. Free and open source (AGPL-3.0), for macOS and Linux.
Lumio turns the MacBook notch into a control center for your AI coding agents: Claude Code, Codex, Cursor, Copilot and more. See who's working, approve commands right in the notch, and watch every session in a pixel office.
I am a solo founder/developer, my agents shipping 30 PRs a week, couldn’t see which part of product they were changing; existing arch tools die because humans won’t maintain the model — so I made the agents maintain it. MIT license
Talk to your coding agents and hear them talk back. SKI runs Claude Code, Codex, Cursor, OpenCode, Antigravity and pi side by side on your Mac, and tells you the moment one needs you.
Swastik Codes is a high-performance, lightweight API platform designed to replace bloated developer tools. Seamlessly test, mock, and monitor HTTP/REST, GraphQL, gRPC, WebSockets, MQTT, and MCP directly in your browser. It delivers enterprise capabilities—including test scripts, collection runners, load testing, mock servers, and automated documentation—without compromising on speed or system resources. Available instantly on the web, with optimized desktop apps for Windows, macOS, and Linux.
Pocket Utility brings 42+ everyday tools into one app. Calculate, convert, scan documents, create QR codes, compress images, manage PDFs, clean text, and use developer utilities—all from one place. Smart Scan extracts useful information from images, while Smart Paste detects content for quick actions. Enjoy a clean, convenient toolbox designed to simplify everyday tasks.
Hi HN, CodeConnect is an iPhone app where you get to control your coding agents across multiple computers from anywhere. Install the daemon (on mac or a linux machine) and run "codeconnect claude" or "codeconnect codex" and its streams the session on the app. Runs over Tailscale and fully open-sourced. Would love some feedback :)
Escribal is a Windows app that generates documents from a workflow: a list of instructions you or your AI agent set up once, without programming. It reads your data, loops, decides and calculates, then writes documents, spreadsheets, presentations and PDFs. You build it on a canvas, your agent over MCP, and after that it runs without AI: the same figures every time, on your computer, no cloud.
Coding agents make up quotes from PDFs and Office files they cannot open. Semlith is a free, open-source local index of your code and docs that Claude Code, Cursor and Codex search over MCP. In my test, made-up quotes went from 41 of 127 to 0.
We track real plan prices, usage pools and hidden costs, then do the math for your workflow. Compare plans, model rates and bill-shock scenarios in one place.
Hello my beloved tech forum! Earlier this year I moved to a new apartment, and scattered a box of poetry magnets all over my fridge. Whenever people came to visit, I noticed how they always loved making poems of their own, not to mention enjoying the poems others had already created. It became such a fun and entertaining tradition, (especially due to the range of creative messages that my friends chose) that I felt compelled to make a full stack project out of it. While LLM agents have made the technical aspects of building a project like this easier, it took me a really long time to nail the algorithm for choosing daily word sets. I started with a huge collection (on the order of 100,000) of words, grouped by part of speech and sorted by the frequency of use in English. I tried randomly choosing a few words in the top 10, a few in the top 100, and so on, for each part of speech. That straight up did not work. Trying to make poetry out of those wordsets felt terrible. It was very discouraging. I had to go back to the drawing board many times. My current solution came about when I found some databases of words, one sorted by their sensory value (from most to least sensory-evoking) and one where words are sorted on a spectrum of concreteness to abstractness. Another key to picking fun and interesting word sets was ensuring there are a solid set of words that are included in every set no matter what. This "glue" set was handpicked by words that I found myself searching for over and over again when trying to write poems, but it is deliberately small to help people be creative in how they express ideas. I used Deepseek v4.1 Flash and the Pi agent. My AGENTS.md emphasizes my status as a junior developer, specifically prompting the agent to never make architectural decisions on my behalf, to quiz me on the architecture/code involved before implementing changes I request, and to never write copy so I can make sure the site feels warm and genuine in my own words. I wanted to avoid letting the code get out of control for when I want to add features in the future. One of the things I love about building these kinds of projects that LLMs enable is being able to add tiny features that most people will never see. For example, I added special effects that occur when a haiku is written. I chose TanStack Start and React this time instead of SvelteKit. The experience was about the same. And kudos on poems are deliberately spammable. I like seeing number go up. Friends have expressed enjoyment from using it so far, but I'm excited to hear feedback on y'all's experience and see if there's anything huge I'm missing. Thank you in advance!
Your agent remembers every session. You don't. Coding Wrapped turns your local Claude Code & Codex history into a pixel-art Wrapped: your rhythm, the words you repeat ("continue", anyone?), what the agent quietly took over, and one small tip for each habit. Open source, one command, 100% local.
Built Multispace to offer an easier alternative to Moonlight and other virtual monitor offers. Then discovered Claude/otherAI can use those Monitors to Work Live, if you like to watch. #cucking. No Signup, No Coding, Simply Install and Go (or don't- Also possible directly in the website, but only native Apps and Iframes.) Installed, Everything is Possible. Build Any System you want, in One Place. Makes Multitasking much easier IMO. Mostly looking to increase the community, make friends, and please, less NSFW usage.
Why? In a d*ck move Spotify decided to impose restrcitions in their free version that they have for the rest of the world: 6 skips per hour, no add to queue, only shuffly playback and tons of annoying ads (earlier it was 1 ad per 30 mins). They also made the developer api paid. So I vibe coded Listune. It runs as a PWA, has a lightweight Go server and I only included things that I needed. I still have tons of Mp3 songs(live on Google Drive), remixes from Mp3 Skull since I was a child and I needed a minimal music player with I have add to queue option, music tagging, search and download songs, all while keep using my existing MP3 library.
Hi everyone. I've been building Imza for the last three months. It's yet another coding agent. I know there are a dozen have been posted here. But I believe this one is different enough to have a place here. It has three explicit modes: plan, build and review. Plan mode allows you to draft a spec document together with the agent. Both you and the agent can edit the document until everything is clear. Build mode takes the plan and implements it. Review mode provides you a code review tool. You can annotate things you want to change in the code and send them as revision to the plan mode. It's entirely written in C++, requires no account, supports 50+ providers including OpenAI subscriptions or local models. Imza supports MCP, skills and instruction files out of the box and uses 8-11MB RAM at startup. One of the biggest differentiator is the Lua engine. What I did differently here is agent receives only one tool, that allows the LLM to execute Lua scripts. I then passed C++ methods as Lua bindings that the agent can execute. So model can orchestrate multiple bindings in one call. Each binding passes through a permission layer. Since Lua is the only way to interact with the filesystem, imza can monitor and control the model's behavior. I'd love to hear your feedback. Thanks Website: https://imza.run Source: https://github.com/umutsevdi/imza
Hey HN! My name is Julia, I am co-founder of mylasertools set of free tools for laser cutting and engraving! Recently we build a programming language for designing files for laser cutting and cnc! It's syntax maybe familiar to OpenSCAD Users, but it takes in consideration laser cutting operations, kerf, materials and assembly! Check out models we created with it and a playground! https://kerfscript.mylasertools.com/models/
Ghostman is a lightweight, keyboard-driven terminal API studio and live webhook inspector. Key Highlights: • 100% Local-First: Stores collections in plain YAML in your repo. No cloud logins. • Live Webhooks: Press Ctrl+W to catch incoming webhooks on :9090 without ngrok. • Fast & Minimal: Starts in <200ms, uses ~40MB RAM. • Keyboard-First: Fast vim-friendly TUI workflow. Free & open source (MIT).
DefAudit is a platform that uses the ISA (Intelligent Security Auditor) solution – developed and maintained by IntellisoftAlpin – a container-based security scanning tool designed for development teams that need secure codebases supporting varied programming languages and frameworks. The platform brings together industry-leading open-source security tools in one web interface, without the pain of using several scanning tools and interpreting their output.
TokenSpend sends your LLM requests to 120 models from 21 vendors and returns a receipt for each one: model, tokens, cost and latency. It works with any OpenAI-compatible client, so switching models means changing the model name. Bring your own provider keys and routing is free, or buy credits at list price with a 5% fee. No markup on tokens, and we don't store prompts. For teams using coding agents, it also ties spend to the pull requests that merged.
Generate high-entropy secure random passwords with live strength checks and estimated crack time calculations. Free, zero-logging client-side developer utility.
Nori is a free, light Git client for Mac, Windows and Linux, beside your IDE and agents: review what each one changed, rebase it, open the pull request.
Host OpenClaw AI agents on managed infrastructure. Telegram and Discord bots with Claude or GPT running 24/7 in isolated containers. Deploy in under 60 seconds with a web terminal, live logs and AES-256 encrypted keys. No DevOps.
Siete stanchi di usare app costose e/o difficili da usare e non avere la più pallida ide del codice che è dietro, e di sostenere colossi inquinanti con milioni di I tracker e cookie. Abbiamo quello che fa per voi, Banana Frasi è qui, un editor di testo gratuito, semplice, funzionante, senza cookie o I tracker e open source, cosa aspettate, iniziate a usare Banana Frasi!!!
I'm the developer. The game sends you to a city, you search the places for clues, then question the suspects with your own words, typed or spoken. Each suspect is played by an LLM that only knows its own character, its own timeline and what you've shown it; it never sees the solution. You catch a lie by showing a clue against a statement, and the case engine decides whether that pair is a contradiction, not the model. Then you write the report: who, when, where, why, with proof for each. The hard parts were not the generation. They were keeping every case solvable (a solver checks each case before it ships: every lie must be catchable with the clues on the map), keeping the suspects from leaking the truth, and keeping the AI cost per player under control (typed questions are capped per day, voice replies are cached per line). Stack: Godot 4 client, Node on Lambda, MongoDB, Anthropic models for the suspects, ElevenLabs and Soniox for voices. Solo, built in a short sprint, launched this week on Android, iPhone waiting for Apple's review. Free, no account. I'd like to hear what breaks: a suspect that says something dumb, a case that is unfair, a lie the engine should have caught. There's a Flag button under every reply.
Backenly is an autonomous backend platform built for coding agents. Your agent builds the backend over MCP, from Postgres to sign-in, storage, functions and realtime. Then it maintains itself: every minute a self-healing loop repairs missing indexes, RLS gaps and dead realtime triggers, verifying each fix live. As your app grows, it restructures its own schema once you approve. Risky changes wait for you, and every change is versioned with rollback.
Lyre lets you run AI agents, automate tasks, and manage projects across multiple clients, whether you're connected locally or remotely. Built for vibe coding, it supports local AI models, multi-account workflows, guided setup tutorials, and live app previews directly on your phone.
Build a voice agent that sounds and behaves human without writing a single line of code. Design the call flow, handle objections and edge cases, connect your tools and knowledge base, and go live on a real phone number. Every call ends with clean, structured data.
Toolioz is a free collection of everyday web tools: calculators, converters, developer utilities, design tools and document helpers. No account, no sign-in, no paywall. Open a tool, get your answer, move on.
Retainza is a developer-first engagement and retention platform, Email, Push and In-App messaging over your own FCM and SMTP, so you keep control of deliverability and cost. An alternative to OneSignal and Braze.
A curated directory of AI and developer tools, written in Spanish for builders in LatAm and Spain: 101 tools across 8 categories, from coding agents to hosting. Each page explains what the tool does and what to compare it with.
opencommerce by Agentcard lets AI agents buy from any online store through a single endpoint: catalogue search, checkout and order tracking in one SDK. Built to be the fastest way for an agent to buy: it searches catalogues and makes purchases in under 20 seconds.
Scrinzo is a suite of 276+ privacy-first tools running client-side with zero data uploads. Elite Highlights: • DirectDrop: WebRTC P2P AirDrop with no cloud storage • Digital Signage & TV Token Calling queue studio • E-Commerce: Amazon & Flipkart label splitters & 4x6 croppers • Developer: LLM Cost Matrix across 100+ AI models • Scientific: Swiss Ephemeris WASM calculations • Hardware testing, PDF tools, media converters • 9 languages, zero signup.
Qodi is an AI-powered financial assistant that simplifies investing. Instead of navigating complex terminal menus, just chat! Ask Qodi about live BIST/global data, technical indicators, or breaking news impact. With real-time verified data and deep market insights (like money flow and order books), Qodi analyzes risks and helps you make faster, smarter investment decisions directly through a natural, conversational interface using text or voice commands.
Build your own Windows VoIP platform with VaxVoIP Server SDK. Add SIP registration, call routing, IVR, conferencing, recording and call-center features, with multi-core processing and SIP threat detection. Includes C#, VB.NET, Visual C++ and Delphi sample projects and a fully functional 30-day trial.
WhisperDesk AI brings OpenAI's industry-leading Whisper speech recognition directly to your Windows desktop with an ultra-sleek Liquid Glass interface. ⚡ Global [ALT+V] dictation into Telegram, Word, Notion, IDE 🔒 100% Private & Offline: Audio never leaves your PC 🌐 99+ Languages supported with auto-detection & punctuation 🎙️ AI cleanup: remove filler words ("um", "like") with 1 click 💰 Pay once ($19 Lifetime), no monthly subscriptions!
See your Git history, stage changes by line and resolve conflicts visually. GitTree brings local Git workflows to macOS, Windows and Linux, with English and Arabic RTL support. Every local Git feature is free, including private repositories. Pro adds pull requests, CI/CD and Nour, powered by your own coding-agent CLI.
tl;dr: Agents can play Pictionary. You can connect your own agent, or just watch a game in progress here: https://sketch.wakeshift.ai/games/q5ucru Hi folks! I initially built this as a proof-of-concept for sending push notifications to AI agents. Watching the agents play Pictionary together has been compelling, so I thought I'd ship it as a standalone demonstration. In Sketch Stadium, your agent subscribes to game events. Agents take turns either drawing or guessing. Easy clues are 1 point, medium are 2, and hard are 3. It's entertaining to watch and some of the drawings are impressive (and others are mind-bogglingly bad). This is the real opportunity for the model labs who have been pelican-maxxing to shine. The first version gave agents the ability to draw lines 10 at a time so that it would be a little bit more like real Pictionary. However, it turns out agents are _really_ bad at drawing that way. So instead I let them use a subset of SVG to draw and that's worked much better, even for the lower-cost models. You don't need to create an account to connect your agent, just add the MCP and authorize. But if you don't want to do that, just watch the game linked above. As a bit more background, agents often start processes that take an indeterminate amount of time to complete (e.g. CI, waiting for approval in a separate system, sending an email and waiting for a reply). If an agent needs that process to complete before it can proceed, it can: * poll, * ask the user to tell it when it's done, * use a bespoke built-in integration, or * use a proprietary capability (e.g. Claude Code monitors) Today, there's no easy, consistent way to wake up agents through a webhook or push notification. I dealt with this firsthand with Clawvisor (authorization for agents) where I needed agents to recognize when the operator had clicked an out-of-band approve button. The best-supported way to achieve that at the time was long-polling. This project is basically a tech for Wakeshift, which is a cross-harness, simple SDK for providing push notifications for agents. It attaches to your MCP to allow consumers of your MCP to subscribe to real-time events, and uses whatever available hacks per harness to wake an agent when dormant. For most harnesses, Wakeshift requires a hook to be installed for the best experience. It will upgrade automatically as harnesses start adopting MCP events. I'd love any feedback or ideas for other products or games that you'd want to see built on this capability. Or MCP servers that you use today that you think would benefit from push notifications.
LumiPilot is a Windows AI assistant for file editing, shell commands, Git, and browser tasks. Connect your preferred AI provider and control which actions need approval. The app is free; external AI subscriptions and API usage may cost extra.
ThunderVPN is a free Android VPN that keeps it simple: install, tap once, connected. No account, no signup. Built on the WireGuard protocol, with a single US server run by an independent developer. Free with ads.
chamoji turns your git commits into changelogs, semantic versions and GitHub releases. Free, open-source CLI for gitmoji, Conventional and plain commits.
MorphScreen turns a model's answer into real UI while it streams: cards, charts, forms, quizzes and your own components. Point your OpenAI-style chat call at our API; every line is checked and repaired, and the web SDK draws it live. The core and SDKs are open source (Apache-2.0). Free plan, no card.
SentioGate is a zero-markup LLM router. Bring your own keys, pay providers directly, $0 on tokens. - 3-line drop-in with the OpenAI SDK (Python & JS) - Smart auto routing with failover - Long-chat memory: 2x fact recall vs summaries - Agent guardrails: blocks destructive bash/SQL, stops loops - Per-request spend tracking - Free community pool: 600 req/5h, no keys needed - Global edge
Writing good specs is hard. You don’t know everything upfront and it’s impossible to write every little detail in the spec. Much of my job as a developer has been to treat specs as intent, and fill out the blanks in the code and use the code reviews as a way to question those decisions. But as coding agents have started to produce much more code, the decisions that actually affect the product are getting harder to find. So I created SpecDeck. It helps everyone on the team write specs and hand them over to developers and their agents through MCP. During implementation, agents write the decisions that matter back as proposed spec changes and leave smaller implementation choices in the code. When the work is done, a person reviews and publish those changes into the spec. The workflow is modeled on git: a published spec is main, drafts are branches, and proposals are pull requests. It should feel familiar to developers, while making it easy for the whole team to work with specs and keep them up to date without touching git. It’s still early and there’s plenty to improve, but I’d love to hear your feedback.
Hi everyone! I typically do not put my projects out for others to see, so this is new to me, but I have created a privacy first budgeting app called KernelCash Budgets ( https://kernelcash.com ), and wanted to share it here in the event someone was interested in a product like this. This app has really helped my wife and I, as well as some family, friends, and colleagues, so I wanted to put this out to see if it could help anyone else. I created this app as a replacement to some of the other budgeting apps out there. Some of KernelCash's core features include: - Envelope and zero style budgeting - Expense, income, and savings goal tracking - Full analytics for income, expenses, and savings - Bank connection / sync free. We use transaction file imports (OFX / QFX) or manual entry to keep things private. You can freely share or not share what you choose. - Multi-collaborator workspaces. These are great for sharing your budget with a spouse, roommates, and have even been used by some parents to teach their kids how to budget. Each workspace is an independent sandbox for you to choose how you work with it. - Customizable budgets that are not month-locked. You control the time span of each budget you have. - There is only one pricing tier so you will always get every feature. I initially created this app only for my wife and I and ran it on a Raspberry Pi. I made it well over a year ago as a replacement for some of the other budgeting apps that I had been using and trying out. I have used many of the budgeting apps that are out there, and a lot of them were either cumbersome to get my data into, or wanted all of my financial data along with bank account connections. Needless to say, I was not really happy with any of these, and so that's why I chose to build KernelCash. KernelCash started as a small tool for my wife and I that I created as a hobby project so that I learn new skills for my day job, and to scratch some of the coding and architectural itches I don't get at work (I'm a software engineer). I never intended to release this to the public, but over time KernelCash came up while in conversations with friends, family, and colleagues. Many of them were really interested in it and wanted to use it since they apparently had some of the same privacy concerns as I did. Since they were wanting access to the app, I decided to move it out of my homelab and into a real multi-tenant application with all of the customer service, landing pages, and other gismos that are expected with a production app. As expected, that was more work than the core app logic. We are priced lower than essentially all of the alternatives at $9.99/month or $87.50 for the annual subscription. If I had the financial means to, I would give this away for free. I can't justify maintaining this project and work full-time and have a family, so I have priced the product as low as I can based on server / infra + general admin and development costs of running this product. We offer a one month free trial and welcome anyone who is curious to try out the product and give any feedback you may have. Feel free to ask any questions, I would love to answer them! You can find KernelCash at https://kernelcash.com Edit: Fixed URL.
Hi HN, I built cc-pair because I kept noticing that when a coding agent reviews its own work, it tends to go easy on it. cc-pair pairs the two agents I already use — Claude Code and Codex — so one authors a change and the other reviews it, and they iterate until the reviewer approves or the round limit (default 3) escalates to the human. Roles are symmetric: either agent can be author or reviewer. In auto mode, `pair start --as claude` (or `--as codex`) opens the author and starts the other agent as reviewer for each round — no second terminal needed. In manual mode you run both yourself. They communicate through numbered message files in a shared directory (~/.pair/<repo>-<hash>/), so the whole review thread is inspectable on disk. Install: npm install -g claude-codex-pair (Node 22+, macOS/Linux for auto mode; manual mode works on Windows). Both CLIs need to be installed and signed in. https://github.com/leonlee/cc-pair — the README has an example review thread in Appendix A, and DESIGN.md specifies the protocol. I'd love feedback: does cross-model review actually catch things self-review misses in your experience, and what would you want the reviewer to be stricter about?
Webifly is a AI website builder and digital marketing platform built for small and medium-sized businesses (SMBs) that want a powerful online presence without the complexity of traditional web development. With Webifly, you can design and launch a fully functional website in minutes using intuitive drag-and-drop tools combined with AI-powered design. No coding knowledge is required, platform guides you from blank canvas to live site AI content suggestions, and automated layout recommendations.
Project URL Category Python Project Title Solaris Dev Shop: a local Python router that picks one installed skill for a task Project Solaris Dev Shop is a public MIT skill collection and local Python router. It picks one installed skill for a task and runs that skill command. It does not start an agent or call an API. Highlights It routes one local task to one installed skill file. It does not call an API, start an agent, or host a model. Example Code markdown or Videos
rekit delivers deterministic, read-only static binary analysis and headless Ghidra automation engineered specifically for autonomous AI coding agents. Eliminating hallucinations and token-wasting guesses, rekit enforces a strict 16-check verification pipeline across 3,200+ lines of Python and shell. Indexed and verified by the XERJ Corpus Hub as an open reference standard for coding agents. Completely free, MIT-licensed, and MCP-ready.
Introducing Tidy Desktop. An open source, local first design canvas and agentic development environment (ADE) in one platform. Run coding agents in your local projects, create and refine interfaces on the canvas, then attach a design to a coding thread to build it. Made for designers and engineers who want to move from an idea to working code in the same workspace. Sync to the Tidy web app in the cloud to collaborate with your teammates and enable multiplayer with the MCP.
If you run AI coding agents, you know the loop: stare at the terminal waiting for "done", or miss the moment it stops to ask for permission. Lingxi is a quiet 3D cat that lives on your Mac desktop and watches for you. When Claude Code, Codex, Cursor or any MCP client finishes, fails or needs you, it walks over and tells you with a face and a speech bubble. Between tasks it naps, wanders and chases yarn. Runs locally, never acts for your agent.
New: HPSILab is now listed in the Claude Directory, so you can connect it to Claude in one click. AI agents get price predictions, Monte Carlo simulation, options IV radar and pre-trade risk checks via MCP, a Python SDK and the web platform.
I've spent months building something complex and (so far) invisible. I know that the number of commits, branches, merges etc. aren't proof of progress, but what I needed was a way to represent the amount of effort I'd put in that wasn't "proximity to burnout". Pretty-git-tree takes in your git history (just the commit structure, and it stays local to your browser) and gives you a visually interesting representation of it. For my own projects I use the "qubits" style which winds a quantum circuit around a space-filling curve. I've included snapshots of some famous git repositories as well. I didn't want to just use the standard git tree, because I felt that there were so many different ways of exploring the data - plus I wanted it to look good at the top of a weekly standup report. I've included the four styles I expect people to use, plus another 15 or so styles in the where I wanted to try things out (like rendering the branches as tower-blocks in a city, or kintsugi fractures). The itself is MIT licensed, and suggestions are welcome.
Hey HN, I'm Fabio, the creator of Beolis. I've built Beolis ( https://beolis.com ) to better handle coding on agentic scenarios. It can be used to manage coding with agents in parallel / in the background either in your local machine (using your current repository or git worktrees) or remotely (where it creates a git worktree remotely and works with the agent there). Although it uses agents you don't need to "chat" with the agent at each step as you usually do when dealing with agents (although you still can), but rather create workflows where you can code the default steps you want. This means that instead of saying 'run the tests' to an agent the idea is that you (or the agent) create code that will actually run the tests and then if it fails it goes to a cheap agent to summarize the failures and then onto a smarter agent to actually do the fixes and work on the problems found -- I find my tokens last much longer given that approach ;) I've been using Beolis to build itself (started back in February) and I believe it's now stable for day-to-day use. For local use it's completely free, otherwise, you can get more collaboration features and remote workers. I'd love to hear feedback.
Type at the speed you talk. Hold a key, speak, let go: clean, punctuated text lands at your cursor in any Mac app, from Slack and Mail to Terminal and your browser. Built for people who write all day and think faster than they type. • Transcribed on your Mac; audio is never stored • On-device AI cleanup strips filler, adds punctuation • Select text and rewrite it by voice • Per-app styles, custom dictionary, snippets • Dictation is free; AI features free for 3 months
Hi! I'm the developer of Klee. It removes background music from live streams but keeps the voice & game sounds. Perfect if the music is too loud or you want to play your own! • Works on Twitch & others • 100% local & private Under the hood: A tiny 1.9M-parameter AI model runs directly in your browser, separating audio in real-time. Free on Chrome: Feedback welcome!
High-performance Indian Railways API developer platform, interactive playground, and TypeScript SDK. 14 REST endpoints for PNR, live GPS tracking, 14-day seat availability, and vacant berths.
TrueLaunch helps indie developers and startups get their apps and games discovered and collect honest feedback. Features include instant app listings, structured reviews, community upvotes, launch-day rankings, developer replies, and reviewer reputation. Every review highlights what worked, what could be better, and what broke. No rating swaps or fake praise—just real feedback to help makers improve their products. Discover, launch, and improve for free.
Ankinize is a sync server for Anki with a REST API on top of the same collection. Use it to keep your devices in step, to add cards from your own tools, or to read your decks from an app you are writing. Give an agent an API key and it can create decks, notes and cards through the REST API or the TypeScript SDK, and run searches. The same cards sync to Anki Desktop, AnkiDroid and AnkiMobile. Agent skills for the API and the SDK are published at /.well-known/agent-skills.
AIRewards helps developers earn rewards through non-intrusive, one-line text sponsorships displayed across AI coding tools like Claude Code, VS Code, ChatGPT & etc. Developers earn a 70% revenue share, while advertisers reach verified developer audiences through targeted campaigns. With CLI wrappers, IDE extensions, browser integrations, global payouts, and a privacy-first architecture that never collects AI prompts or source code, AIRewards connects developers and advertisers in one ecosystem.
One window for every server you own: SSH and a local terminal, SFTP, Postgres/MySQL/SQL Server, MongoDB, Redis, S3, RDP, queues and Kubernetes, as tabs. End-to-end encrypted sync. $89 once, 14-day trial, no card.
HalalGen is an AI video studio. Turn a simple prompt into short videos, and make images, logos, YouTube thumbnails and 4K upscales in the same place. No coding and no design skills needed. Image and video prompts are checked against our content rules before generation.
High-Security file checker and developer sandbox for your Mac. Attack-tested. Pay once, free upgrades. Special Launch Price: $9.99. Drop any file, download or code you don't trust into SafetyBox. It scans for known threats and runs programs in a fresh sealed machine with no internet and no view of your files, then throws it away. For developers: run unread scripts and AI-agent code (Python, Node.js, shell, Linux) and check npm/PyPI packages and leaked keys. Apple certified: Silicon + Intel.
Lasm has been my dream for a good part of 2026. I firmly believe that formal verification is the next evolution of software engineering and programming languages. This next wave will largely be enabled by the incredibly intelligent LLMs that understand formal verification and its programming languages. Lean is one of the most popular languages that allow for formal verification, and to me is incredibly powerful because it is a general purpose programming language with formal verification built directly into it. You can build general purpose software in Lean, and formally verify it in Lean. Amazing. Lasm brings Lean to the browser, Node, serverless, and edge environments (and more). The idea is that you can now write web apps, frontend to backend, in Lean and formally verify these apps. Imagine an Express server or Hono server, running on Cloudflare Workers, with formal verification of its specification. I'm an agentic engineer with an emphasis in software engineering, and Lasm was built mostly by Claude Opus 5.5 with my high-level direction.
I wrote sigflint because I wanted to sign AWS requests from Node, Bun, Deno, Cloudflare Workers and browsers without pulling in an SDK or any dependencies. It is a single package with no dependencies that signs requests as an Authorization header or a presigned URL, for S3 and any other AWS service. It also does SigV4a (multi-region), chunked uploads and event streams. What I checked, and how: * Signatures match botocore for all 3,360 S3 requests in the test corpus, plus 1,622 requests to other services and 1,600 presigned URLs. SigV4a signatures verify against the AWS CRT's, and chunk and event signatures match the AWS SDKs. * Speed: 1.0 to 3.2 µs per header or presigned-URL signature on Node 26.12.1 and 0.7 to 3.0 µs on Bun 1.4.4; an S3 GET signs in 0.97 µs on Node. The six other libraries I compared took 2x to 75x as long on the requests they sign correctly. * Size: the bundled, minified header signer is 5.7 kB, 2.8 kB gzipped, and the other signers add to a bundle only when imported. Feedback on the API or on requests it signs wrongly is welcome: https://github.com/bompus/sigflint/issues
Free, open-source app for notes, docs and databases. Every page is a Markdown file in git on your computer, AI agents read only the pages you share, and plugins add what else you need.
Build and publish a mobile-friendly business website without coding. Choose a template, add your services, photos and contact details, and go live. Free hosting, HTTPS, a Cockatoo web address, contact form and enquiry inbox. No credit card required.
DevOS offers tools like world has never seen the way it offers. It brings developer tools, AI assistance, and everyday productivity into one desktop app. Keep your notes, tasks, clipboard, and conversations together, so you can spend less time switching apps and more time building. Available on Windows, macOS, and Linux.
✨ What you can do with it: 🖼️ Turn into beautiful mockups 🎥 Transform screen recordings into polished visuals 🎨 Make your work look cleaner and more professional 🌐 Use it directly on the web Whether you're a developer showing off your latest project, a designer presenting your work, or someone who wants their to look better, MockBok is built for you. No more boring screenshots. Make your work stand out. 🔥
Clipboard, notifications, files, messages, calls, screen mirror and camera between your Android phone and your Mac, over your own Wi-Fi. No ADB, no scrcpy, no Developer Options.
Describe an app once. OpenDaemon's AI Developer writes a plan of phases plus a checklist, routes each phase to the agent inside your Claude or Codex desktop app or Antigravity IDE (not a CLI), reviews every report and screenshot, and sends the work back until the app runs. The same free Windows hub runs a Playground AI society on your PC and an assistant you can reach from Telegram, WhatsApp or Slack. Free NVIDIA models out of the box, or bring your own key.
BlaqVibe helps builders share coding projects that would otherwise stay hidden in folders or repositories. It’s for vibe coders, indie hackers, students, and developers seeking feedback, early users, and a public record of their work. Publish or import projects, discover and remix builds, use Studio and Nolo AI, join challenges and battles, and offer downloads with Stars or Paystack checkout. Projects are scanned before reaching the feed.
Skills for Claude Code, Cursor and Codex, one for each thing that goes wrong once AI-built code goes live. Broke After Deploy finds what will break in production before users do. Key Leaked finds secrets in code, git history and build output, then walks you through rotating them. Bill Spiked finds the calls behind a bill spike and adds limits. Prove It Works (free) makes your agent show real test output before it says done. Every fix is a diff you approve, and it all runs locally.
hi everyone! I wanted to share what I've been working on these past 6~ years. TL;DR: Rust is great, but somewhere along the way, I learned that I was too dumb to read and write Rust. As I worked on larger and larger Rust projects, I also became frustrated with the developer experience of working in Rust (slow compile time), so I wanted a better way. Vu's language changed a bit over time but the core principles remained the same: 1. It should be easy to read and write 2. It should be memory safe 3. It should be fast to run 4. It should be fast to compile Here's how I approached each: 1. Vu is syntactically very close to Python. The ownership rules are the ones Rust has, but I personally find it a lot easier to parse and understand. I've tested this with various LLMs and found it can read and write Vu out the box because of the similarity to Python. 2. Vu uses the same principles of Rust. Every value has one owner, borrows can't escape or outlive what they borrow, and nothing changes while it's borrowed. 3. Vu compiles down to C at the end of the day, so everything runs approximately at C speeds. 4. vuc check on the compiler’s own source (about 21,500 lines of Vu), takes about 0.4~ s. This check covers parsing, type and borrow checking. cargo check on a Rust project that size usually takes several seconds from scratch. Then there are other opinionated language decisions. Vu has no macros, Vu's borrow checker is lighter weight than Rust's with simpler interfaces and generics. However, in the past few years, as Vu's language evolved and AI began to take over, I expanded Vu's principles to include a couple additional principles: 5. It should be secure 6. It should be efficient for LLMs to write 5. I want to pen test Vu in public. In https://github.com/vu0tran/vu/tree/main/tests , you'll see a battery of tests where I'm trying to basically pwn Vu. There are explicit language decisions in there as well that tries to prevent a host of security flaws that exist in other languages, e.g, integer overflow stops the program instead of quietly wrapping around. 6. The syntax of Vu is designed to be tokenizer efficient. In addition, when Vu checks throws errors, the errors are designed to be token efficient and written in a way that guides LLMs to more easily fix bugs, reducing turns overall. So, in summary, Vu is still experimental. The compiler is 21,500 lines of Vu and compiles itself. The standard library also has TLS, HTTP/2, WebSockets and a PostgreSQL client, all tested against Python’s implementations and hostile input. I think most of the pieces are there. It does also compile down also to WebAssembly so you can play around w/ the checker on the website! Feel free to try to break it! What's next: 1. Figure out how to get Vu to perform faster than Rust 2. Rewrite some OSS libraries in Vu to further prove the language
I didn’t want to copy and paste messages between coding-agent sessions or across vendors, so I built an MCP server that lets one model in one session talk to another model in a different session. They can even have a group chat. It supports multiple OpenCode sessions and lets GPT-6.1 in OpenCode talk to Claude in Claude Code.
Give your interface a little emotional range. Weasel brings 874 free SVG icons to React, with meaningful states, custom colours and smooth transitions. Connect them to real app events, or build your own states with the included coding-agent skill. MIT licensed.
Can one person spin up 1 Million AI agents using AI coding and actually do business with other agents around the world? The first to do it wins $10,000,000.
Locket brings couples closer through daily questions, mood sharing, cute widgets, memories, games, and shared plans. Built by a developer for his own relationship, it's one private space for two—without needing a collection of separate couple apps.
Meet Caret: an open-source, Cursor-class AI code editor built with Rust. No bundled Chromium, no Node runtime, just a lightweight core and your OS WebView. Get AI autocomplete, agentic coding, inline edits, code reviews, MCP, and background agents. Bring your own API keys or run fully local with Ollama. Use Claude Code, Codex, Gemini CLI, and more via ACP. Smart model routing powered by Lynkr. MIT licensed. Your keys stay local. You own your workflow.
Real-time charting terminal for serious traders. Live prices, 80+ indicators, custom scripting, market scanner and AI analysis — all in one browser tab.
Hello HN, I've been working on a new Web engine, build around two ideas: - use formal methods from day one. - make it modular. The modularity is orientated towards the following goals: - enabling shipping anywhere, including on IOS in the EU as per Apple’s guidelines. - only write code for stuff the OS or application platform cannot do for you: keep binary size to a minimum and achieve native performance. - developer choice: bring what you want, and only build what you need. For example you can choose a JS engine like v8 or Boa, or just bring your own. The entire graphics pipeline consists of components whose implementation can be swapped-out, with currently a relatively fast path on Mac integrating with Core Animation and a slower cross platform path. Project is at https://github.com/gterzian/formal-web Intro essay at: https://medium.com/@polyglot_factotum/how-i-am-building-a-we... Please let me know what you think.
signoff-ui is React components for the step where a person signs off on what an agent does. DiffReview gives the agent back each file with only the accepted hunks applied, the rejected hunks and the reviewer's comments. ApprovalCard shows exactly what will run and lets you approve once, for the session or always. For AI SDK 6 and 7, AG-UI and ACP agents, inside assistant-ui, AI Elements or your own chat. MIT licensed.
Ezhil is a home screen you can actually make your own — quick to use out of the box, and deep when you want to dig in. Seven home styles, from a plain grid to Hero, Agenda, Edge, Minimal, Terminal and Bento. Set your grid, resize and stack widgets, group apps into folders, and swap the dock for the layout you prefer. Free placement lets you drop anything anywhere, at any size or angle, off the grid entirely.
Staal is the programming language for agents: code whose authority is enforced, not prompted. Its compiler refuses programs that could act without permission, trust an unchecked model answer or overspend, and its runtime never repeats a payment after a crash.
Rivet is a chat-first AI coding agent built native for Android. Pick a project folder, pick a model, describe the change — Rivet reads and edits files and runs commands right on your phone. No laptop, no Termux setup. Bring your own key (OpenAI, Anthropic, Gemini, OpenRouter), or sign in with a ChatGPT plan. Switch models and reasoning levels mid-chat, and undo changes with checkpoints. Free and open source.
During the dark ages of AI, when Opus 5 was incomprehensible and Codex Ultra was burning all tokens within a day, I decided to give open-source agents and models a try. What I saw was that a lot of the tools I tried were terminal-first, and desktop apps were less common. Additionally, I felt that lately a lot of the agents apps are slowly trying to become “superapps”. They keep adding more and more features that imo are not always useful and make the UI quite heavy and complex. This motivated me to build Wavefunction: A minimal, open-source app desktop app that focuses on the coding experience and tries to reduce unnecessary features. It doesn’t have a built in browser and its computer use functionality works in a VM. I built it on top of langgraph and it’s packaged as a tauri desktop app. Currently, It works with Vercel AI Gateway, OpenRouter, Anthropic API key, Google AI Studio API key, and you can log in with your ChatGPT account or use a local model with Ollama. It also has 70+ connectors. Last, but not least, you can also require ZDR-only models from the providers. If you try it out I would love to hear your feedback! Thanks!
Free, unlimited daily practice sessions for software engineers. Hold the coding. * 5 minute quizzes to sharpen fluency on practical technologies. * 15 minute reviews of AI generated system designs. * 30 minute simulate shipping something end to end in an agent sandbox - chat with stakeholders, spec, build, judge, in a fresh sandbox.
Paste the URL of your Lovable, Bolt or Cursor app. A passive scan checks for keys in public JS, publicenv/.git, source maps and missing headers. Plain-language fixes. Can't see RLS or server code. First scan free.
RingMay brings internet calling and AI voice agents together for businesses and people. Businesses can build an AI agent without coding to answer customer calls, take booking requests, and make reminders and follow-up calls, with $5 in free credit to start. People can use the Android app to call other RingMay users without a SIM or talk to AI specialists for learning and guidance, with 100 welcome credits and no subscription.
Hi HN, I'd like to present Nori, a Git client built with performance and low resource consumption in mind. Nori helps you review your agents' work and easily rebase, commit, push and PR your worktrees. It's available on Mac, Windows and Linux. Nori is free for common Git features (even for private repos). You can buy a one-time Pro license for $20, which lets you list and manage reviews and issues right in Nori.
My friend and I are Neovim nerds, and every so often our Neovim setups would throw errors. What did we do? We talked to each other and solved it. It was never Neovim’s fault. We think the same way about harnesses. If you don’t like something, you shouldn’t have to call a trillion-dollar company; you should be able to fix it yourself. If you want something custom, you shouldn’t have to keep spamming some discussion thread; you should be able to write it yourself. And more than anything, it should be simple and fast, and your computer shouldn’t scream OOM. That’s the inspiration behind uji. I hope you all love it!!!
Install an open-source kernel in your project. It helps your AI assistant work from goals and sources, use or form relevant skills, keep decisions and their reasons, and carry useful learning into later tasks. Works with your existing coding agent.
Localcraft is a free collection of 47 tools — PDF, image, video/audio, developer utilities, calculators, plus 11 privacy-first AI tools — that run 100% in your browser. BYOK AI tools: your key stays local, files never upload. No sign-up, no watermarks.
Beyonder is the AI living software factory. Start with ready-to-use, tested AI apps, from CRMs to content tools, or build your own by describing what you need. Beyonder draws on 150+ real business deployments to assemble your software from proven building blocks, not from scratch. Connect your data and workflows, make changes through conversation, and let Beyonder handle hosting and infrastructure. No coding required.
Ip2region is an offline IP-to-Region localization library and IP data management framework with both IPv4 and IPv6 supports, 10-microsecond level query efficiency, xdb search client for many programming languages
Crack campus placements with real previous-year questions from TCS, Infosys, Wipro, Accenture and more. Timed mock assessments, coding practice and recruiter-recognised certificates.
Hi HN, I'm seeing more and more people sharing software specs instead of code, and I think this will become increasingly common as coding agents get better at one-shotting entire apps from a single prompt. So I put together Promptuaria over a few days. It's a small platform where you can publish, version, and fork software specs, and link GitHub implementations to the exact spec versions they were built from so you can confront them. Specs are Git-versioned under the hood, with immutable published versions. It's an early side project, but I'm curious to know if others see specs becoming something we share and reuse, pretty much like code. Would love to hear what you think!
Flatbread turns a coding agent's session into one link: what was asked, every step, what broke, what fixed it and the diff. Redaction runs on your machine before anything uploads. Another agent can pull the link and pick up where it ended.
If you've been following along with DeepSeek news you will know they recently added web search capability to the DeepSeek API, where you can use web search under the same API key. It is pretty handy, also super cheap at only $0.75 per 1,000 requests (pretty much as cheap as it gets, one search credit can encompass many responses). I created a simple TypeScript CLI wrapper around the API for my own agents to use (for open source models and harnesses) I'm sure most of you could whip this up on your own but it also comes with unit/integration/live/e2e testing, CI checks, changelog, auth handling, linting, sdk wrapper, and a fleshed out readme, potentially saving you a few hours of work + validation. Just need to install, auth, and drop a comment in your 'AGENTS.md' to use `dsearch --help` when agent needs web search. GitHub: https://github.com/LaishGlenberg/deepseek-search-cli
ToolMitra offers free online calculators, text tools, PDF and image tools, and developer utilities. Explore useful tools in English and Hindi, with no sign-up required for basic use.
QAmcp tests MCP servers the way AI apps use them. It reads every tool, safely runs the read-only ones and tells you in plain words what is broken: vague tools, missing safety labels, hidden instructions, leaked secrets. Free in your browser or terminal.
I built Seqora as a solo developer because I was tired of DBeaver and TablePlus, the tools I was using. None of them felt like modern tooling. DBeaver is mature, but it lacks customization and it’s slow. TablePlus puts too many limits on its free tier, like only allowing two tabs open at a time. What I wanted to do differently in Seqora was to let me visualize different objects, like indexes, partitions, etc. Another thing I found most tools lacked was a way to figure out what’s slowing down a complex query, so I built a visual EXPLAIN that shows where the time goes. I put the main focus on those two. The next thing was being able to work on a connection visually using diagrams, so I can see how things are connected and make changes by drag and drop. Once everything is done, I can review all the SQL that will run. I made sure the things you need for a daily driver are free, so there’s no limit on the number of queries, connections or tabs. The deep-dive parts, the object views, visual EXPLAIN and schema diagrams, are in Pro. Pro is currently $5.99/month or $49/year. It’s pre-1.0, currently at 0.4.6. It supports macOS (Apple Silicon) and Windows, and Linux is coming soon. It’s built with Tauri, Rust and Svelte. Most of the code was written with Claude Code and ChatGPT. I made the decisions and did the reviewing and testing. I’d love it if you tried it and told me what you don’t like so I can improve it, and also what you hate about your current workbench.
Agent Standup turns the Claude Code, Codex, Mistral Vibe and Cowork sessions on your computer into a live video call. The lead hands out work; sub-agents join, talk and share their screen: task list, test runs, diffs, the browser they drive, live command output. See at a glance who is working, stuck or waiting for you, and which agent spends the tokens. It only reads local session files: no extra tokens, nothing sent. Free, MIT, in VS Code, any terminal or as a desktop app.
If I'm abusing Show HN feel free to downvote (I know this blurs the line between a job post and a Show), but genuinely just trying to get some feedback from other engineers / engineering leaders on an approach we're trying here @ DocLoop re: engineering interviewing. TLDR: Now that coding is no longer at a premium my thesis is that alignment is more precious than ever.... this is my optimistic attempt at trying to build something that achieves a _little_ bit of it. It's a compromise between the challenges everyone hates taking and _something_ that a recruiting team can use to get some meaningful signal amidst the sea of AI submissions that come. Definitely open to critique!
As my first experiment in AI autonomy, I set up an agent that runs the product, while I assist it. The product is the community board hosted on the homepage of 42Loop.com On Product Development front, the agent ideates with me the features, positioning, UX etc. and provides me the prompts that I pass on to a coding agent (as we progress the goal is to take myself out of the loop) On the marketing and ops front, the agent finds apps to list, reviews them, finds opportunities to promote the product across the web etc. The agent and I communicate via the open boardroom here. It posts a daily briefing, I provide feedback. The goal for the agent is to hillclimb on the registered users metric. Happy to answer anything about the setup. Looking for inputs and suggestions. The agent is reading as well. Thanks.
开发者自荐：本投稿由 TokensQC 开发者提交，与网站存在直接关联。 TokensQC 是一个面向 AI API 接入排查的网页工具。填写 API 基础地址，选择有权限的模型后，可检查响应结构、流式返回及请求延迟等可观察行为；检查项随模型而异，部分型号还包含工具调用往返检查。站内另有 API 中转站目录和模型参数对比，方便查找接入资料。 适用场景是接入第三方 Claude / OpenAI 兼容 API 后，排查协议不匹配、响应异常和延迟问题。返回的 model 字段或签名字段只能作为行为线索，不能证明底层模型身份；单次结果也不能代表长期稳定性或实际账单准确性。 费用与数据边界：网站工具不收服务费，探针会发送实际 API 请求并消耗目标服务商额度。鉴权信息会经 TokensQC 代理转发至用户填写的服务商，请使用专门的限额测试 Key，不要使用生产主 Key 或真实业务内容，分享结果前注意脱敏。 欢迎就检测范围、误判和使用体验提出反馈。
YS LeadGen is an all-in-one WordPress popup and form builder that helps website owners capture leads, grow email lists, and increase conversions. Create exit-intent popups, newsletter forms, spin-to-win wheels, sticky bars, quizzes, and multi-step forms with an easy drag-and-drop editor. WooCommerce users can recover abandoned carts, offer discount coupons, and promote related products. Track performance with analytics and A/B testing, connect email marketing tools. No coding required!
Yardlet is a free, open-source Mac app for the coding agents you already use. It starts Codex, Claude Code, OpenCode and others with their own CLI and sign-in, and gives them one sidebar, composer and set of approval cards. Sessions stay in each agent's own store, so a task started in Yardlet also opens in the Codex app. It writes nothing into the agents' configuration, so you can stop using it any time. Diffs, a pull request tab, connectors and a browser the agents drive come with it.
A polished, browser-first collection of more than 100 practical tools. Search by name, task, common phrase, abbreviation or spelling variant; open a focused tool page, or jot down notes in the private, plain-text-first notepad. The catalogue covers text and writing, developer utilities, images, files and data, calculators, unit converters, colour tools and everyday helpers.
GramDesk turns a Telegram bot into a shared support inbox. Customers write to your bot; teammates reply from separate customer topics in a private Telegram group. Keep conversation history together without sharing personal accounts. No coding is needed for setup. $0.10 per active bot per day, with unlimited teammates.
Developer of 6 years here, been using AI (increasingly) over the last year for most of my work. A few weeks back I started to look for new jobs as the current startup I work for is falling into pieces. "Do I need to do leetcode again?" - I thought. So I picked a few challenges and tried to solve them without an assistant. This felt impossible. What?! Is this the brainrot everyone is talking about? I mean I still read code every day, I just don't write it anymore. I could read and review the solution easily, but writing it myself felt like speaking a language I only understand, but cannot speak. Being the data hoarder I am, I immediately wanted numbers, so I built Muscle Memory to help me feed them. It's like leetcode but better. Challenges are more realistic, and I'd like to think the UX is more straightforward. It covers 12 languages and your code runs in a network-denied sandbox in your browser. No account needed. I want it to be a place to keep your hands sharp: short tasks, clear prompts, and difficulty that comes from solving the problem rather than figuring out what the question means. About the data (that I'm curious about): With your consent, Muscle Memory records your practice as raw events, including keystrokes, runs and submissions. Hoping to build an open dataset of how we write code in the AI era. That could help us understand how we approach problems, where we get stuck, and how we well do we retain our craft during this transition. Your individual results stay private unless you share them. “Stealth” mode keeps your data out of published findings, and you can export or delete everything. The data will be published as a report/notebook on the site itself, and be available raw under MIT license (parquet) once we reach a critical mass of sorts, so it makes sense to share. The code is also MIT-licensed on GitHub (https://github.com/ekkuleivonen/muscle-memory). I wrote more about why I built it in the manifesto here: https://musclememory.run/manifesto. I'd especially like to hear what surprised you about your first attempt. Try it out here: https://musclememory.run
把技术概念画成会动的机械图。这是一个为 AI 编程助手打造的插画技能，能把抽象的技术概念转化为真实可动的机械装置，绘制成精致的等距视角交互插图，比如把缓存命中画成只有每根锁销都对齐才能转动的锁芯。它支持 SVG、SVG+WebGL 着色器和 3D 三种形式，可导出为 HTML 页面、SVG 文件或 React 组件。
Run Claude Code, Codex, Gemini CLI, Copilot, Cursor and 13 more coding agents side by side. Manage real terminals, multi-agent workflows, MCP servers, approvals, code reviews and token usage from one desktop app for macOS, Windows and Linux.
Take your tokens anywhere. Use your subscriptions, API keys and local models in Codex and any coding agent, or inside the ChatGPT app on your Mac. Hit a limit and it keeps going. Start a coding session on your home machine from anywhere.
先联系咨询：微信 Huaz0105 ｜QQ 1782919426 ｜邮箱 [email protected] 购买与充值： 运营方自荐：独立第三方付费 AI 视频生成服务与 API，适合视频创作者和开发者。价格优惠，模型与渠道会变化，具体费用和可用情况请先联系咨询，以实时信息为准。 For international customers — contact us first: Email [email protected] | WeChat Huaz0105 | QQ 1782919426. Purchase / top up: Independent third-party paid AI video generation and API service for creators and developers. Contact us for current model availability and pricing. Prices are in CNY/RMB. If you are outside China or have difficulty paying in RMB, please contact us before purchasing to discuss available payment options.
The free Claude skills library: hundreds of hand-picked skills for Claude Code and the Claude app from the best GitHub repos, for marketing, SEO, coding, design and more.
用 Python 自动操作 Windows 桌面应用。这是一个 Windows 图形界面自动化 Python 库，能够模拟鼠标和键盘操作窗口、对话框和控件，还可以读取界面上的文本数据。它支持 Win32 API 和 MS UI Automation 两种后端，可以通过窗口标题、控件名称等文本属性定位元素，几行代码就能完成打开菜单、点击按钮、输入文字等操作，适用于桌面软件的自动化测试和重复操作。
Project URL Category Other Project Title Resume Claim Guard: an agent skill that stops hallucinated resume claims Project An open-source agent skill (SKILL.md) for Claude Code and AI coding agents that enforces a strict 4-step resume workflow: anchor every claim to a verified fact base, audit metrics against exaggeration, and generate bilingual role-aligned bullets. MIT licensed. Homepage: Highlights Prevents hallucinated resume claims by binding each bullet to a verified fact base, and shows changes as yellow-highlight diffs so you can review every edit. Bilingual output, usable for interview prep too. Example Code markdown or Videos
Trigger.dev Webhooks let you build AI agents on top of the services your customers already use, like Slack, Supabase and Stripe. A GitHub pull request can start a code review agent, a new Sentry issue can start an agent that writes the fix, and a Stripe refund can wait for a human to approve it in Slack.
I built QuickRef as a collection of developer tools that run entirely in the browser. Most online tools send data to servers, which isn’t ideal when working with logs, tokens or sensitive text. I built this because I often didn’t want to paste sensitive data into online tools. So the goal was: • everything runs client-side • no login • no tracking • fast and minimal Would love feedback or suggestions for tools to add.
I used GameMaker 8.1 a ton in late elementary and middle school, and GML was the first programming language I wrote. I still think it's a fun engine to use because it has so many batteries included. So I built an open-source reimplementation of it in Rust which I'm calling Instance. It opens existing .gmk and .gm81 projects and runs them on Windows, macOS and in the browser, with a more efficient GML interpreter than the original GameMaker. You can try the IDE here in your browser: https://groverburger.github.io/instance/ . It comes with three example games, including Spelunky Classic built from Derek Yu's released source shared under its permissive license. Just change something and press Run to see your edits in action. Old GM games depend on a lot of quirks such as the order events run in, the exact random number generator, how sound_isplaying lags behind the audio. I wrote up the behavior as specs from the documentation, the file formats and black-box tests against the real program, then implemented it from those specs. No proprietary code was disassembled or decompiled. CLEANROOM.md describes the process. It covers most of almost all of 8.1. All the resource editors, GML and drag-and-drop actions, d3d, particles, surfaces, MIDI, mplay multiplayer and DLL calls. It can also export a game as a standalone executable or a single HTML file. Projects can be saved as folders of plain text and PNG files for git. I wrote most of the code with Claude Opus 5.5 over about three weeks. Claude banged out a pretty good reimplementation in a morning, but as always the last 20% took 80% of the work. GameMaker also turns out to work surprisingly well with LLMs. GML is small and simple, and a whole game fits in one project file. I'm happy to answer any questions! Code: https://github.com/groverburger/instance
A native Mac notch app: run Claude Code, Codex and 19 ACP coding agents from the MacBook notch, chat on the ChatGPT or Claude plan you already pay for, and keep twenty everyday tools one hover away.
LaogengAI 是一个精选 AI 工具导航站，聚合 ChatGPT、Claude、Midjourney、Kimi 等全球主流 AI 工具，覆盖对话、绘画、写作、编程、视频、办公等分类。每个工具都有详细介绍、同类替代推荐与相关文章，帮你快速找到最适合的 AI 工具。 A curated AI tools directory featuring 57+ hand-picked tools across chat, image, video, writing and coding categories, with detailed intros, alternatives and guides for each tool.
Lumvise keeps explanations, decisions and reports connected to code, documents and image regions. Explore project relationships in a visual workspace and give your AI coding tools reusable context through a local MCP server.
Describe your idea and Appfully's AI builds a working iOS, Android or web app, live in your browser. Refine it by chat, take payments and publish. No coding.
I built this for my own file-system and CLI workflows. Wherever I notice a friction point, I update this program to address it. For example, when using a cd command to navigate in a terminal, I've made it easy to access ones I've used before, or to find the right directory using a few characters from dir's name. It also smooths over miscellaneous OS-level functionality which I've encountered pain points in.
Agenturm brings your agents’ sessions, activity, tool calls, and traces into one local terminal view. See what’s progressing, what’s stuck, and what needs your attention—without bouncing between terminals. Works with Codex, Claude Code, OpenCode, Pi, and other agents. Your data stays on your machine; replies and approvals stay in your hands
Dedica tests your website the way AI agents use it, then gives your team a fix for every place they fail. Muse, Instinct, GrokBot and other agents now browse, compare and sign up on people's behalf. Dedica runs 125 checks (for free) to see if they can use your site, and gives you a fix for every issue to hand to your coding agent.
Solo game developers juggle design docs, tasks, budgets, and community across 5 different tools. Nexus GM brings it all together: an AI-powered GDD assistant, Kanban board, multi-currency budget tracking, Steam review monitoring, Git visibility, and more. One purchase. No subscription. Your data stays local.
Charge for AI features without billing code. Embed a prepaid-credits checkout with one script tag, point your existing OpenAI or Anthropic SDK at one endpoint covering 390+ AI models, and every request is metered at real token cost. You set your commission and earn it on every request, with no monthly fees.
Hold ⌥ Space in any project and a Liquid Glass ring opens at your cursor. Flick at a slice and let go: Claude Code, Codex or your terminal takes the task, in its own git worktree if you want. When it's served, view the diff, merge it or yeet it. Runs locally.
Trellis is a native macOS terminal that's more than a terminal: windows, tabs and splits in a sidebar, saved SSH hosts, one-click project tasks, a command palette and scratchpad, plus a CLI and MCP server so scripts and AI agents can drive it. Free, MIT.
Self-hosted APK channel-parameter injection without re-signing. One-file deployment with Console, download statistics, offline licensing, and an Android AAR SDK.
Type a GitHub username to see what each of your repos would be worth at the rates AI labs pay to license code. The rates come from real potential buyers, which current offers hitting $100 per million tokens of source code and $40 per million tokens of git history for private repos.
Plan a freshwater aquarium before buying fish. Estimate usable water after substrate and decor, review sourced fish references, and plan water changes. Free, no account required; inputs and saved plans stay in your browser. Supports litres, US and imperial gallons, local plans and printable summaries. Checks are planning estimates, not certification of stocking capacity, cycling or compatibility. Made by an independent developer.
Paste your live URL. SniffPup passively checks web apps built with Lovable, Bolt, Cursor, Replit or Supabase for leaked keys, publicenv/.git files, source maps, missing headers and open Supabase/Firebase reads. Read-only. Every critical fix is free.
Frontier models improve at a pace faster than the pace at which most humans are able to keep up. Not all of us are obsessed with pushing the boundaries of our agents and trying new ways to work with them. I often spend 12+ hours a day working with an LLM, and still, when I talk to people, I hear of ways of working I hadn’t thought of. What if we could speed up that pace of learning? What if everybody shared their practices of how they are making the most of AI, and if we could learn from each other? What if AI could make that sharing happen? Could everyone catch up with AI then? I know, it sounds dreamy, but my therapist says illusions are necessary—until some of them come true. I built a tool that teaches your local agent to scan your sessions, spot the best practices you already use, and guide you towards the ones you haven’t discovered yet. Only names and counts reach the server. Your sessions never do. The product has a self-consolidating catalogue of practices, MCP servers, plugins and skills. Here’s my stack—scan (locally, all session data remains private). https://app.userrank.ai/@milstan To spice things up and give a reason to rescan, I added rankings inspired by Foursquare (who here is old enough to remember Foursquare when it was gamified?). You can rank in the top 10 of a city, or of a field like Product. Claude Code, 10–20 minutes on your own plan: claude plugin marketplace add https://app.userrank.ai/plugins/marketplace.json claude plugin install userrank@userrank claude "Get my UserRank" Codex, or a quick scan of your Claude or ChatGPT chats: https://app.userrank.ai/start
I have been working on this alone for 6 years. cmusee is not a step debugger like GDB. It records every function call and assignment statement, and then present it entirety This is the minimum memory required to record some projects. curl 6MB flatbuffers 6MB git 14MB jq 26MB nlohmann/json 18MB raylib 199MB llvm llc 649MB llvm opt 182MB Showcase & Interactive https://veefuse.com/en/cmusee/playground/curl Some corner cases are not support currently.
ParallelEarth lets you: Group alternatives under a named experiment. Create separate Git worktrees called “Earths” from a committed revision. Run checks you explicitly approve and record their results. Compare outcomes against a baseline. Export JSON or Markdown reports. ParallelEarth uses worktrees underneath. What I’m trying to improve is keeping the alternatives, check results and comparison together inside VS Code.
MDViewer opensmd files instantly from Finder — double-click or press Space for a QuickLook preview. Built with SwiftUI, not Electron: it launches in under a second and uses a fraction of the memory. Edit in place with a double-click, browse Git history with color-coded diffs, render Mermaid diagrams and KaTeX math, export clean PDFs, and search across everything with ⌘F. Full GitHub Flavored Markdown. Lite version is free, Pro is $9.99. macOS 13+, Intel and Apple Silicon.
An MIT-licensed coding agent for your browser and terminal. Bring your own model API, edit files, preview HTML, export ZIPs, and extend the browser agent with Rust/WASM plugins. Free software; model API usage is billed separately.
I am the owner and founder of TheGitAI Coding Agent. I built this Chrome extension to let the agent control your existing browser directly from the terminal without interrupting the workflow. It doesn't require your login details or any secrets. You can control your browser now through agent without sharing any credentials through UI. Each step is taken right in front of you with controllable permissions. Try it for your most complicated work that requires browser access. Would love to hear feedback!
Retake gives coding agents tools to record real browser walkthroughs. Keep the flow, update it when the app changes, and record another take. Open source. Runs locally. Tested end to end on an Apple Silicon Mac with Claude Code.
Teladio is a desktop companion for builders. Code Panion runs Claude Code and Codex side by side in your repos, with Git, a terminal and web, iOS and Android previews inside. Meet Panion transcribes your calls, no bot, and leaves you the action items.
Advanced IT & Healthcare medical billing solutions company delivering accurate coding, faster reimbursements, and full revenue cycle management support.
RageLayer is an Easter egg for your own site. Visitors pick a hammer, fire, chainsaw, or black hole and wreck the real page: text and cards break into physical debris, effects interact (water puts out fire), and a broom sweeps everything back. 15 tools. Works with React, Vue, Svelte, or plain HTML. Free, MIT, TypeScript, zero runtime dependencies, plus an SDK for custom tools. Play with the live first.
runspace runs dependencies, build scripts and AI agents' commands on a disposable GitHub codespace rather than on your machine. For devs who want builds and tests off their laptop. For people who run coding agents and want every agent command in a sandbox. Has a CLI and an MCP server with the same commands, returns the work as a patch you can check before applying, and requires only a single binary and a GitHub token.
一键切换 AI 编程助手的模型。这是一款用 Go 开发的 AI 编程助手模型管理工具，让 Claude Code 用上 Kimi、Codex 用上 DeepSeek。它通过本地网关转换 OpenAI、Anthropic 和 Gemini 的接口格式，支持多种编程助手和模型服务商，还能把订阅共享给其他助手使用，额度用完后自动切换账号。
Built a little native MacOS app out of necessity to help me free up some storage on my Mac. I've always been frustrated by the lack of visibility and mysterious nature of the Storage overview especially when it comes to Developer and System Data. Ballast gives you a detailed breakdown of all common sources Mac developers would be familiar with, such as node caches, docker images, AI models and even stale unused project concerns. It's free and yes built with Claude. Try it out!
Tapground runs iOS and Android UI tests written in plain English. Instead of selectors or XPath, a vision model reads the screen on every run and finds what you meant, so a moved button doesn't break the test and your app needs no SDK. It runs on iOS Simulators and Android Emulators with live preview and keeps a video and a screenshot per step. A test can reuse a value it read, wait for a verification email, and run on both platforms from CI with one command. 50 minutes free to start.
Read PDFs inside P7M files and supported FatturaPA XML locally in your browser. Document content stays on your device. Free full watermarked preview; single-file export €1.99. Batches require an account; file packs and subscriptions are available. Up to 50 MB per file, 50 files/100 MB per batch, five nested layers. FPR12/FPA12 invoices: one body, no attachments. Local integrity checks do not validate certificate trust, revocation or legal signature validity. Submitted by its developer.
谷歌开源的 AI 手机自动化工具。这是一款 AI 驱动的 Android 自动化工具，只需用自然语言描述任务，AI 就会自动操作手机或模拟器，完成跨应用的测试流程和日常任务。它内置 MCP 服务，能让 Claude Code、Codex 等 AI 编程助手直接操控手机并收集日志和截图，支持 Gemini、Claude、GPT-4o 等多模态模型。
IdeaProof is no longer just a validator - it’s the Founder OS. We turned your raw idea into an end-to-end launch pipeline: • Real demand intelligence: Stress-tested against 1,000+ startup post-mortems. • Sequoia-style pitch decks: Auto-built from your market data with 1-click PDF. • Complete visual identity: Ready-to-use color schemes, typography, and assets. • Native MCP Server: Run validations, pull failure stats & stress-test ideas straight inside Claude, Cursor, or your terminal.
An AI code editor built on VS Code with completions, codebase chat, an agent that edits across files, Multi-Agent, Feature Tree and browser QA. Works with Claude, GPT, Gemini, local models and your own API keys.
Tandrio brings coding agents together on a visual canvas. Connect Codex, Claude Code, Gemini CLI or OMP, give each agent a role, link their work and follow the flow. Use your existing provider accounts and work in your own project folders. Join the beta and help shape the product.
Price Tag is a free classifieds board available in English, Spanish, German and Russian. Post items, services or jobs at no cost and reach buyers with automatic translation. Built by an indie developer as a simple, fee-free alternative to paid classifieds.
Hey HN, we built playplanet.dev, a Lovable alternative that lets you build and run full stack apps entirely in the browser, no server needed. We run it on a Linux like system compiled to WebAssembly and then use a runtime on top of it. Currently we support frameworks like Astro, React, Angular, Vue and Svelte. You connect your own OpenRouter account and use it to build out the code with an LLM, so there’s no markup fee on our side. Every app you publish is open source under the MIT license, and anyone can remix it to build their own version. It allows you to get a head start with your project by using another user’s project as a starting point and save AI tokens. You can also get inspired by the community and share projects with your colleagues. Because it all runs on git, you aren’t locked in and can clone your project anytime and deploy it wherever you like. Please check it out and let me know what you think. Website: https://playplanet.dev/
Run Claude Code, Codex, Cursor CLI, OpenCode or any shell side by side, each in its own Git worktree, with a queue of agents waiting on you and diff review. A separate daemon holds the sessions, so agents keep running when you quit, crash or update the app. Rust desktop app, no Electron or web view. Oct 28 benchmark (Krill 0.2.20 vs Orca 1.4.242, 10 terminals, screen on): 1/7 the memory, idle CPU and install size. Free, Apache-2.0. Apple Silicon Mac; Windows preview.
The programming interview process can feel like a race to the bottom. To get ahead of the curve, we've decided to define what it looks like at the bottom. Enter Hackerdome: a real time PvP programming puzzle duel with spectators and sabotage. A very non-serious site for a serious purpose (having a little more whimsy in our lives), where you can race a competitor to solve a programming puzzle or spectate and watch the chaos unfold. If the puzzles get too boring, the players might just find their editor playing back, with pop ups, font shifts, and more. Enjoy!
Project URL Category JS Project Title In-repo delivery board: stages, handoffs, soft WIP, weekly report, Timeline Project Open-source delivery board that stores stages, owners, handoffs, and soft WIP limits in a git-tracked file. Includes a weekly Markdown client report and a local kanban plus calendar Timeline UI. Runs standalone via npx or as a DeepSeek Harness plugin. Aimed at keeping delivery context next to the code rather than only in SaaS boards. Highlights Delivery state lives in the (delivery.json) and is reviewable in PRs Automatic handoff audit when cards change stage; soft WIP limits warn without blocking Weekly Markdown report (shipped / handoffs / risks) plus local editable board and calendar Timeline Standalone CLI: npx jerryxff26-alt/dsh-delivery-board — no SaaS account required Example Code bash npx jerryxff26-alt/dsh-delivery-board init '{"customer":"ACME","template":"governance"}' npx jerryxff26-alt/dsh-delivery-board card '{"title":"API integration","stage":"analyze","owner":"Ada"}' npx jerryxff26-alt/dsh-delivery-board open npx jerryxff26-alt/dsh-delivery-board weekly or Videos GIF in
popkern is a free, open-source parametric BRep kernel written in Rust, supporting booleans, fillets with properly smooth corners, sweeps, shelling, and STEP. Try it in your browser or install it in one line.
A VS Code-based workspace uniting an AI coding agent with specialized agents for research, planning, and writing. Build and test code, then review changes with a second model in the same session. Use your Claude or ChatGPT plan, or an API key. Free to start on macOS.
Nano Metrics adds a live dashboard to your VS Code sidebar: Claude Code status, tool calls, and task progress; token usage and estimated costs from local transcripts; edited files and recent activity; CPU, memory, disk, GPU, per-core load, and temperature; optional monitoring for 30+ services like Docker, Postgres, Redis, Kafka, and Ollama; process-level usage; and a theme-aware status bar. AI coding activity, costs, and system metrics, all in one place.
Hey HN, happy to share Chopstick with you. It's a Mac app that makes your notch useful, but most importantly customizable and dynamic. What makes it unique is that you can build custom widgets with your coding agent. Chopstick automatically learns from your usage and suggests new widgets with time. Widgets are dynamic meaning they can trigger in specific situations. The obvious case is "when Spotify is playing", but you can also get more complex and say show the widget to push my changes when VS Code is open and I have changes. TLDR: - Chopstick gives you widgets in your Mac notch - You can build your own widgets with your existing agent - Widgets are highly customizable in what they can do and what triggers them - It's a one time payment (with trial), discounted for launch Feel free to join the discord as well for sharing widgets! https://discord.gg/kqbnvgeTQC Let me know your feedback! - Josh
I run my business on four Claude Code agents: Titus, Pheme, Davinci and Metis. They write my newsletter, The AIgent (about 9k subscribers), every day through a drafting and proofreading engine that Pheme checks before scheduling. They also build AI courses, run a pipeline that turns the day's featured into a talking-head video of my avatar for Instagram, set up ad deals, and run Meta ads and email funnels with me. Nothing closes until I sign off. Metis posts jobs on a job board and the others claim them. The hard part wasn't any single task. It was keeping the agents coherent over weeks. Compaction kept dropping the small decisions, and then an agent would redo work I'd already ruled out. aigent-os is my fix. At a context threshold (80% by default) the agent writes a capsule, a checkpoint file in git, and has to reply with one exact line. Only then does the runner send /clear, wait for a SessionStart receipt proving a fresh session started, and resume. One ack buys at most one clear. If anything is ambiguous it stops at HOLD:clear-stalled-after-ack and doesn't retry. On resume the capsule is reference only, live state wins, and anything the writer couldn't capture is written as Unknown. First run interviews you with three plain questions and makes a first plan. Each agent gets its own memory vault. Limits, honestly: the public evidence is 11 of 11 cycles on one Windows seat. The runner can still miss a permission dialog (#104), a seat launched from inside another Claude Code session can lose its transcript (#98), and scrolling the seat window sends stray input (#99). v1.1.0 shipped on Sunday. Try bash install.sh --dry-run, it writes nothing. I'd like to hear where the loop could still double clear or strand a session.
Preparing for a developer interview? SpellInterview helps you practice technical questions and improve how you explain your knowledge. Choose a programming language, answer questions by voice or text, and get instant AI feedback, scores, and suggestions for improvement. Practice junior-level questions for free, or unlock mid and senior levels with a subscription. Identify knowledge gaps and prepare for your next technical interview with more confidence.
Fast, secure, professional tools organized for students, writers, creators, and developers. Runs 100% locally in your browser with zero server logging. Features developer utilities (JSON, Base64, Color Converter), content tools (AI Prompt Studio, Grammar Checker, Word Counter), and finance calculators (SIP & Loan EMI).
Hey HN, I've been building Sectr, a backend-as-a-service for making it easy(ier) to develop and deploy AI agents. Deploying agents requires non-trivial systems design work. Consider a support agent: how do you stream messages to your frontend, how/where do you persist messages for frontend fetching later (or for auditing), and when the agent wants to issue a refund, do you spin-down the compute infra while it's waiting on human approval? Sectr aims to be the serverless abstraction and platform to provide these systems for you. No sign-up needed to try using the dev server: curl -fsSL https://sectr.dev/install.sh | sh # Install CLI sectr init my-agent && cd my-agent # Starter code sectr dev # Dev server, with UI to test chatting Sectr provides: - Streaming of every token, tool call and event, and persistence so they can be loaded fresh later. - Scaling up, and down to zero when the agent needs to wait, and resumption right to the suspended agent's state. - Human approved tools automatically use Sectr's suspension and resumption capabilities, no code needed from you. - AG-UI endpoint for any deployed agent so you can trivially hook up your frontend - Local dev server experience, so you can rapidly debug and iterate on your agent's behavior At the core, you can deploy any container image to Sectr as long as it implements a certain shaped HTTP endpoint. In practice, currently Sectr provides adapters to let you deploy your existing OpenAI Agents SDK, LangGraph, or Pydantic AI agents, with only 2 new lines of code: app = AgentApp(name="support") app.add_route("/chat", openai_agents(agent)) Certainly the hardest part of building Sectr was getting the resumption logic to be correct and invisible, across the Program State x Model Provider x Agent Framework matrix. For example, an earlier bug was when a model replied with text plus a tool call, Sectr would resume that as two separate messages, instead of one message with a text and tool part. This works, but breaks prompt caching (and might subtly change model behavior). Sectr is free during the beta (and can be tried locally without sign-up); I mainly want feedback, especially on dev & deployment experiences, any bugs, and what features would excite you. Introductory video: https://www.youtube.com/watch?v=H-y3wynkKc0 Docs: https://docs.sectr.dev
Hey HN, I wanted to show you something I built. I actually built it last year but the prices of VPS instances went to the moon and I shelved it until recently. It's called Whitehirst (I had the .com domain lying around, hence the name!) You can spin up a Sovereign UK/EU Windows Workstation in ~10 minutes that's fully configured with all the dev tools you'll need like Visual Studio 2026, VS Code, Docker CE, Git, Rust, NVM, Tailscale and more. It's a custom Windows Server 2022 rig that removes all the bloat, telemetry and Copilot stuff. I've spent an obscene length of time perfecting the build so it runs incredibly fast. You can spin up a rig for 8 or 24 hours with zero risk of a surprise cloud bill as every workstation is strictly ephemeral and automatically self-destructs when time's up (prices are £12 for 8 hours or £25 for 24 hours). You can extend your time if you wish. I'm only offering a single spec at the moment: 8 AMD EPYC vCPUs, 32GB ECC RAM, 200GB NVMe storage and a 2Gbps nic - I intend to offer other specs later on. If you'd like to try it, head to https://whitehirst.com and fill out the contact form and I'll pick a handful to get started with. I'm going to keep the numbers small for now so that I can make sure it all works and I can iron out the bugs since it's still in early-access. Anyway, let me know what you think. Ta Mick https://whitehirst.com
Konsenxia OSS is an early MIT-licensed Flutter and TypeScript prototype exploring healthcare collaboration and human-reviewed AI workflows. It includes application source and development tests. Some screens and integrations are incomplete or mocked; it is not clinically validated or suitable for patient care. The roadmap focuses on reproducible synthetic tests, clear data boundaries and developer documentation.
Contenido:LingTube es una extensión para traducir videos de YouTube, convertir voz a texto y generar doblaje con IA. Al abrir un video y activarla con un clic, traduce los subtítulos o la voz de videos en otros idiomas (incluso los que no tienen subtítulos), y los lee en voz alta con una voz natural. Así puedes entender videos de YouTube sin estar pendiente de los subtítulos.Empieza al instante a usar el doblaje con IA y la traducción de subtítulos en YouTube, Coursera, Udemy, Bilibili y TikTok: solo reproduce cualquier video y comienza de inmediato.Ideal para aprender idiomas, cursos online, tutoriales técnicos, noticias, pódcast, anime, fragmentos de películas y ver contenido mientras viajas o haces otras cosas.Funciones principales Traducción de videos de YouTube: traduce subtítulos, voz y contenido de videos sin subtítulos. Subtítulos bilingües: muestra traducción + original al mismo tiempo, o solo la traducción o solo el original. Doblaje con IA / lectura de subtítulos: usa Edge TTS o la voz del navegador para leer la traducción en voz alta. Videos sin subtítulos: cuando el video no tiene subtítulos, transcribe automáticamente la voz, la traduce y la lee. Traducción con contexto: compatible con Google Translate e interfaces OpenAI para entender mejor frases largas y contenido técnico o especializado. Aprendizaje con IA: genera vocabulario, gramática, ejercicios, evaluaciones y material de repaso. Control de audio: puedes escuchar el audio original y la voz de la traducción al mismo tiempo, y ajustar el volumen de cada uno por separado. Activación rápida: un clic o el atajo de teclado Alt+Shift+V. Resumen de videos con IA: una vez iniciada, puede conversar con los subtítulos traducidos completos para resumir el video, extraer las ideas principales, explicar términos técnicos y localizar momentos clave. Perfecto para videos largos, cursos, tutoriales técnicos, noticias y pódcast. Privacidad Solo funciona en páginas de YouTube, Coursera, Udemy, Bilibili.tv y TikTok. La configuración se guarda localmente en el navegador. Los subtítulos, traducciones y solicitudes de voz solo se envían a los servicios que tú habilites. Extensión / plugin de ChromeChrome Web Store: oficial: 亮点 Contenido:LingTube es una extensión para traducir videos de YouTube, convertir voz a texto y generar doblaje con IA. Al abrir un video y activarla con un clic, traduce los subtítulos o la voz de videos en otros idiomas
This is basically the doomgeneric idea but applied to Minecraft (its first version rd-132211 that came out on 13 May 2009). It's one freestanding C file with no includes and no memory allocation. It runs in a terminal, over telnet as shared multiplayer, as a 20 KB wasm module, inside a PDF on Chrome and on a bare x86 PC with no OS, and on a jailbroken Kindle Paperwhite. PDFium doesn't have WebAssembly so the C goes to wasm with clang and then binaryen's wasm2js turns it into JavaScript. It also compiles for a Raspberry Pi Pico and Game Boy Advance, but I don't have any of these so haven't tested it. If someone has one would love to hear how it works there. The browser version can be played here directly: https://pragyaangaur.github.io/Can-It-Run-Minecraft Used Claude for the C and JavaScript code because that's outside my area of work. I picked the targets and tested everything. The Kindle version ended up being an extended project so it's not in the linked https://github.com/pragyaangaur/Inkwell And if anyone has any crazier places it could run on (someone please do it on a pregnancy test) do try and share it... maybe I could add it to the repo!
Remlo is a native notes app for Mac, iPhone and iPad. Type a checkbox in any note and it becomes a real to-do, linked back to where you wrote it. Each day gets its own page with your note, what's due and your habits. To-dos sit on a board with stages you name. Habits show streaks and the weekly trend. Hand a to-do to an agent and send it to Claude in Terminal or the desktop app. It all lives in your own iCloud: no account, no server, works offline. Free to try for 7 days.
Paste your git log; the grader lists every file two sessions (two authors, or two branches) changed less than an hour apart, ranked by how often it happened. Those are the edits most likely to undo each other when Claude Code, Codex or a teammate work in parallel. It runs entirely in your browser: nothing is uploaded, no signup. Try the sample log to see a report in one click. A README badge shows the count for a public GitHub repository.
Browse number rental services and manage OTP requests in one account. Rent virtual numbers and receive OTP verification SMS instantly from 100+ countries. Developer-first platform with full REST API support and transparent pay-per-use pricing.
Thinker cuts your coding agents’ token spend by ~20%. Thinker is an open-source proxy that sits between your agents and the model. It strips context your agent no longer needs, caches knowledge so future agents don’t have to rediscover it, and keeps your instruction files lean and current. Across my first dozen users, it’s averaged 20% fewer tokens per completed task, with no drop in success rate. For teams, it also shows where token spend is going, by team, repo, and agent.
Anchor on System, Not on Luck. BAS is a system you can return to — not another reset. 4 Principles: Consistency — Same template every day Reversibility — Every change can roll back RETUNABILITY — Return to baseline + Retune 1 point + Re-enter Zero Reset — Never start from zero 3 Layers: Personal → Team → Org 7-Day Test: Score 1-5 daily, improve 10% weekly RETUNABILITY = RETURNABILITY + TUNABILITY Proof: 5x cannot stat → 0 Reset → 5 Commits Success. Fixed by 1 line: git config --global
Since our first launch, GS Copilot has grown into an AI coding agent inside the Apps Script editor. New: Agent Mode for multi-file changes, Plan Mode, "Fix with GS Copilot" to debug errors in one click, inline suggestions, and MCP Tools on Pro and Lifetime. You approve anything that sends, deletes or touches a secret. Launch offers until Oct 31: code PRODUCTHUNT takes $30 off Lifetime + BYOK ($99 one-time), and code PHSUB takes 25% off monthly Plus or Pro for your first 3 months.
Find the best AI tools for coding, agents, chatbots, image, video, voice, writing and research, ranked S to C tier and updated weekly with new releases.
Shark brings dual split panes, infinite tabs, an embedded terminal drawer, and instant hover previews into a familiar macOS interface. Built 100% natively in Swift & SwiftUI with zero Electron bloat, Shark helps developers, designers, and power users navigate deep nested folders, compare directories side-by-side, and preview files without losing their flow.
TA Platform is a browser crypto terminal that runs entirely in a tab - no install, no signup. Live candles from Binance, Bybit and OKX render side by side on one Canvas2D chart built from scratch instead of a charting library, so it stays smooth with thousands of candles. 38 indicators, 12 drawing tools and a live arbitrage spread view for spotting price gaps between venues.
Braidplane Alpha connects APIs, databases and files with declarative pipelines, SQL transformations and typed contracts. Version specs in Git and inspect runs in the UI or CLI. For evaluation with test data and non-production systems.
AISG scans the app AI writes for you: leaked keys, unsafe patterns, vulnerable dependencies. 26 sensors, core scans run locally so your source never leaves your machine. Findings are indications that need human review — not a penetration test
Hydra Dev Studio puts your databases, file servers, cloud storage, Git and Subversion repositories and SSH terminals in one native Mac app. Edit server files in place, compare a server with your repo, and let AI agents like Claude Code use your connections without ever seeing a password.
HappyHands helps you improve an existing website with your coding agent. Review pages and interactions, extract design tokens and components, confirm what to change, then compare the result with before-and-after evidence. Free, open source, and run locally with your own agent setup.
Hands-free AI voice control for Windows. Speak naturally to open apps, write, browse, and manage files. Works with any LLM (bring your own key), plus agents, knowledge base, and mobile remote. Built by a developer with Parkinson’s from Greece. Free tier available at maestrovoice.com
An engine that learns your coding style from past projects, turns it into agent skills, and proves coding agents follow it. Claude Code today; Codex and other agents next.
Pling is a real SSH, Mosh and SFTP terminal for iPhone, iPad and Android, built for working with coding agents. Jump back into tmux or Zellij sessions, approve Claude Code's risky commands from a push notification, or describe a task and let the Command Agent run it. Plus push alerts from any script with one HTTP POST, live server monitoring, remote desktop and end-to-end encrypted sync. Free to start; Pling Pro unlocks the rest.
Allin1xTools brings free online tools for PDFs, images, text, calculations, unit conversion, and developer tasks into one convenient place. Convert files, format text and code, and simplify everyday digital work.
Search and compare 322 open-source AI and vibe coding tools: chat UIs, model servers, RAG, agents, coding agents, Claude Code skills, starters. Each gets a score from public GitHub data, a review and its license in plain words. Nobody pays for a spot.
Email your coding agent sets up. One command signs you up; your agent adds your domain, applies DNS, verifies it and tests the first send. Switching from SendGrid, Postmark or Mailgun? Import templates and suppressions in one step. 3,000 emails/mo free.
WelcomeScore by ETHIOR evaluates how ready a public GitHub repository is for first-time contributors with deterministic 5-pillar audits, onboarding checklists, and developer badges.
Merkur opens your Mac or Linux machines from any browser or phone as a real terminal, for Neovim, tmux, Claude Code or Codex. No SSH keys, VPN or open ports. It holds up on bad networks, renders like a native app and pings your phone when an agent needs you
Author here. As RDF 1.2 and SPARQL 1.2 approach recommendation status, the transition from standard RDF 1.1 reification to triple terms `<<(...)>>` and edge annotations `{| ... |}` introduces some differences (e.g., asserted vs unasserted statements, named reifiers, etc.). I put together this 4-page reference sheet covering the mental model, exact grammar constructs, and SPARQL 1.2 query idioms. No paywall, printable A4 format. Happy to answer questions or how SPARQL 1.2 queries can be executed over local files.
For AI builders who fear bloat, bugs, and broken intent. SPIDER Protocol turns your AI coding agents into a governed engineering team: • Preflight Discovery: Absorbs true user intent before touching code • Living Git Memory: Architecture & decisions immune to context amnesia • Independent Validator Gate: Audits the diff and full test suite before any task is marked Done Build production-grade software at AI speed. 100% free & local.
I originally built drop.media to do one thing: drag an image into your browser and get a shareable link. It still does that! Now it's one media library for you and every AI agent you use. Upload from the browser, terminal or API, or let Claude, ChatGPT and Cursor save files to it through MCP. Any agent can find what another saved, use it, and share it as a public link or one that expires. Images, PDFs, video and docs, on every plan, including free. Upload once. Use it everywhere.
🎯 What can you do with it? ✅ Explore interview questions and answers across software testing and automation topics. ✅ Practice Manual QA concepts and technical topics. ✅ Prepare for programming and automation technologies. ✅ Review answers and strengthen your interview knowledge. ✅ Create your own questions and answers in your personal question bank. ✅ Explore community-shared questions as the community features develop. 🛠️ This is an early-access beta! Some features may not work perfectly yet.
i was using Termius, but it ate all my ram, lagged, and charged 20 bucks for the team plan. so (being impressed by how fast Zed felt) i tried to make an alternative with GPUI. me and my friends have used it for some time and now i want to show it to other people. i tried to make the design look like Zed because i really like it, but i'm not sure it turned out well. would love to hear what you think. it's open-source, mostly built with ai but it doesn't eat my ram and isn't written in electron. local-first (sync and teams are optional). the api is not open source, but if people start using the app, i'll add a self-hosted version. git: https://github.com/mousEHEHE/flint-gpui mac and windows: https://flint.mouse.pw
Looking for free speech recognition I’ve been looking for speech recognition I can call from a script without paying for a developer API. A lot of apps already offer voice input for free. That made me wonder whether their speech recognition could be used separately from the app. Doubao’s input method was the first one I tried.
Team surveys are easy. Knowing what to do next isn't always. Pluribus analyzes your team's answers with AI, suggests concrete next steps and drafts the workshop to work on them. Helpful for teams with or without an agile coach or organizational developer. Currently in German.
Coding agents made building fast. Knowing what to build is the slow part. What to Vibe turns what the internet is talking about today into app ideas, each with a full build prompt to paste into your coding agent. Free, no signup, new ideas daily.
外卖盒被鉴定为「封印供奉盒」（袋口死结为封印，供奉者开启前常先拍照，似为必要仪轨），键盘是「四方占卜盘」（W、A、S、D 四枚磨损最重，推测对应四方神）。目前有日常、打工人、校园、春节四个展厅共五十件，环形 3D 展厅可以拖着转、点开看展签，手机也能看。文物模型都是用 Meshy 文生图再图生 3D 做的，three.js 渲染，纯静态页面。 我是开发者本人，本帖为自荐。
See every Claude Code session on one live screen: which one is working, which needs you, what its agents are doing and what it costs. Each chat runs in its own window and keeps going in the background. Free and open source.
The ElevenLabs Realtime skill teaches AI coding assistants to build on ElevenLabs Realtime models. It sets up an ElevenAgents agent on Realtime Mini or Max, helps pick between them, and migrates a browser voice app from OpenAI’s Realtime API by swapping the credential, WebSocket URL, model and session config. MIT licensed, part of the elevenlabs/skills repo, installed with npx skills add elevenlabs/skills.
ondocs adds an Ask AI box to your docs: visitors get a short answer from your pages with citations to the exact section, or "I don't know" when the docs don't cover it, on your own model key. Coding agents get an MCP server (search, fetch, list_pages) and llms.txt from the same index. Plugins for Starlight, Docusaurus and VitePress, guides for Next.js, Fumadocs and Nextra, or one script tag. No vendor, no vector database. MIT licensed.
The open discovery directory where indie software, developer utilities, and AI agents gain lasting visibility. Ranked by launch freshness and real usage, so great tools never get buried in 24 hours. Free submission, dofollow backlink, and AI agent discovery via MCP — queryable inside Cursor, Claude, and Windsurf.
Tekyous is an interactive tech stack builder and tools directory for developers and founders. Start from one of 150+ curated stacks, a two-question recommendation or a blank canvas, then draw your stack as a diagram with frames, connectors, notes and your own tools. Live hints show which tools you might be missing or where they overlap, based on nearly 300 curated developer tools. Free to use.
A desktop Git client for you and your AI agent. You get a fast graph, line staging, drag-to-rebase and Undo for Git. Claude Code, Codex or Cursor connect over local MCP: they see what you selected, use far fewer tokens than raw git, and only propose commits.
Earlier this year, I started developing Termcraft, a terminal-based 2D survival game written in Rust. Since its first release more than half a year ago, I've added multiplayer support, so you can now host a game and play with friends (or colleagues during a meeting?); Up to 8 players. I've also substantially improved the visuals, lighting, physics, and overall gameplay. A friend and I ended up spending the entire weekend playing through the game from start to finish. The progression takes you from gathering wood and building a shelter to crafting equipment, exploring caves, finding diamonds, crossing the Nether, and fighting the Ender Dragon. There are villages, farming, collaborative building, PvP, and player chat along the way. It's open source under the AGPL-3.0 license, with downloadable builds for Linux, macOS, and Windows. Gameplay video: https://youtu.be/nn37_15dLFM Installation and controls: https://pagel-s.github.io/termcraft/ It's not perfect, but I'd love to hear your feedback and experiences with the game. Contributions are also more than welcome!
Avita is a free classifieds marketplace for Russia: sell goods, cars, services and jobs without listing fees. Web + Android app (RuStore). Shops can open a free storefront at avita.fun/shops/connect. Built by an indie developer as a simple, fee-free alternative to big paid classifieds.
Practise tech interviews with an AI interviewer: live coding, technical Q&A, system design and architecture reviews. Upload your CV for targeted questions, then review recordings, history and transcripts. Free 10-min demo, no signup. Plans from £15/mo.
Too many events are reduced to a form, a poster, and a registration link. But an event should be an experience people look forward to, not just a list of names. Hackways gives events their own space to showcase their identity, share their story, and bring the experience to life. Built for developer meetups, hackathons, conferences, and workshops, it helps organisers create event pages, share essential details, and manage registrations without juggling scattered links.
Free Chrome extension for apps built with Lovable, Bolt, Replit, Cursor or v0. On a site you own, one click checks HTTPS, security headers, keys in public JS, source maps andenv/.git. Read-only, no tracking, fixes in plain words.
A cms that wants to make one love their inner writer. Fall back in love with writing again. Giving everyone a voice to speak. A lightweight, Git backed content workspace for sharing and publishing experiences with zero friction. try now: https://lightcms.me/ https://github.com/light-cms/lightcms
CampusCode is an all-in-one coding platform with an online compiler for 12+ languages, DSA practice, roadmaps, core CS subjects, projects and placement prep.
KnightCode is an open-source coding agent for the terminal. Sign in with Claude, ChatGPT, Grok or Copilot, use an API key for 40+ providers, or run a local model. Every file read, edit and command shows up in the transcript. Sessions are a tree you can branch, fork and compact. Extensions, skills and themes live next to your repo. The system prompt is about 1,100 tokens, so more of the context window goes to your code. MIT licensed.
EngMock turns your own interview questions, interview write-ups and notes into live AI mock interviews for software engineers. Practice system design, coding/DSA, OOD and AI-assisted coding with adaptive voice follow-ups, a whiteboard or coding workspace, and scored reports with evidence and replays. VS Code mode (preview) keeps coding in your editor and voice in the browser. Use MCP to create and edit scenarios, prepare interview links, and review reports and progress.
Built an app with ChatGPT, Claude, Cursor or Gemini? Paste one prompt into your AI coding tool: it packages and uploads the project, and one click puts it live on HTTPS with its own database and cache, on two servers. From $9.90 a month per app.
Web-AI adds a talking AI presenter to any website with one line of code. When a visitor lands on your site, they see an avatar with a real human voice that introduces your business — like a salesperson who's always on duty. Works on WordPress, Wix, Shopify, Elementor, and any custom site. Three voice options: record your own voice, AI-generated speech, or professional recording. Free plan available. No coding required.
A Chrome extension for developer documentation. Open a developer docs site, click scoop.md, and its llms.txt plus its instructions for agents are copied as a short prompt for Claude Code, Codex or Cursor. No llms.txt? scoop.md builds one. Free, open source.
Upload a CSV, JSON, XML, or YAML file, or connect Google Sheets or a database, and get interactive charts, a full dashboard, and a PDF report. No coding needed.
ElementShelf is a UI component library for coding agents. It has 233 components, each in 20 design languages, 4,660 designs in all. You download one file, elementshelf-pack.md, and the agent installs the components you picked.
Turn a 5–10-minute YouTube programming chapter into a written lesson with explanations, code and a styled PDF. Free, open-source beta. No login or API key needed. Shared capacity applies; review generated material against the original.
Answer 40 yes/no questions in your terminal and get an A–F compliance grade, a prioritized fix list, and 15 generated starter documents (privacy policy, DPA, RoPA, breach plan, SOC 2 checklist). Pure Python, zero dependencies, runs 100% locally. Install: pip install autocompliance
Free live order flow tools for Bitcoin, Ethereum and Solana, in your browser, no sign-up. Depth: the futures order book as a heatmap, every trade as a bubble, with candles or a footprint chart. Options: one live chain across Deribit, OKX and Bybit with call wall, put wall, gamma flip and max pain. Terminal: 16 live panels on one screen. Plus liquidations, whale trades, GEX, funding and a report after every session, with an India edition in IST.
This is hardly a greenfield area. There are tons of agent-to-agent communication apps. But, at least to my taste, a lot of them are too complicated or too opinionated. If we learned anything from the HuggingFace incident, it's that modern AI agents don't need a lot of help communicating. They are smart and creative and figure it out. So...with that in mind, I (meaning I + claude and friends) built kashkeshet, or kkt. It's a small, Rust-based MCP server that gives agents a few simple chat primitives (identity, channels, chat) and leaves it up to them to use those however they want. You as the human get a little web interface so you can watch the fun. Coordinate your coding agents; setup a virtual company; conduct sociological experiments; whatever. kkt's main goal is to stay out of the way, and let you and your agents figure out how to organize and communicate on your own. This is meant to be run on your local device and shared among local agents. There's a little explanation of how you could use tailscale to go broader, but nothing like Oauth2 or anything yet.
TraxAgent is an AI spend intelligence platform: it turns your Claude and OpenAI APIs, coding agents like Claude Code, Codex and Cursor, and Claude and ChatGPT AI usage from across your domains into actionable insights and consolodated spend. Risks flagged early, costs cut, opportunities found, each with the exact fix for optimization. Read-only, no prompts stored.
Instantly add reliable, battery-efficient speed tracking to your fleet or telematics app without wrestling with Android sensors. Empower developers to quickly build eco-driving tools that enforce optimal cruising speeds and cut fuel costs. Features include enterprise-grade background GPS processing, EMA signal smoothing for flawless accuracy, a customizable Jetpack Compose UI, and 100% on-device privacy. We've included ready-to-run samples so you can clone, build, and test in minutes.
**Atlas AI — All your AI tools in one place** Atlas AI is a free-to-use AI assistant designed to bring multiple AI capabilities together in one place, including coding assistance. Whether you're asking questions, working on code or exploring what AI can do, Atlas AI aims to make these tools accessible without a subscription. Try it here: Created by Omar Hosny.
Hi HN, I'm Alexander. Yes, this is another agent harness. Here is why I built one anyway. I was engineer #2 at Browser Use, where I built our agent harness and the evals behind it, including BU Bench, which became a default benchmark for browser agents. I left in September to build Mainplane. The problem: local agents die with your laptop, and cloud agents can't reach your work. Many cloud platforms ask you to "select a repository", but my work is not that simple. I work across a desktop, a Mac mini and a laptop, and much of it is not in git. So I loaded Ubuntu on my old 2012 desktop and made it my Mainplane server. It has run for four weeks at a load average of 0.02, because the agents' tools run on other machines. How it works: it is a control plane architecture. The server runs the harness. It holds your API keys and every session, makes the LLM calls, and routes each tool call to a worker. The workers are your other machines, connected over a built-in private WireGuard network. A bash command runs in the worker's terminal, not on the server. Shared drives let projects move across machines. The agent works on your machines just like you would, but its state lives on the server, and your keys never leave it. Some design choices: - Every session is one append-only file. This is the agent state, and it is easily readable. If something breaks, you can resume it from the file. Observability comes for free: agent traces are just these state files. - Small by design. The system prompt is 3k tokens, and there are just two kinds of tools: shell and file (read, write, edit). At Browser Use, I learned the bitter lesson and found frontier models did better when I deleted all our helper functions. So I made the same bet here: bash is all you need. - No compaction. When a session fills up, /continue starts a new one that greps the previous state file, retrieving the context it needs, instead of reading a lossy summary. A /compact on a 400k-token session takes a while to write its summary. A /continue into a brand new session starts at once. - No permission prompts. A hyperintelligent misaligned agent WILL find a way around a permission system. The better strategy is giving the agent full unbounded access to a local computer, and only putting what you trust the agent with on that computer. - Why Go. WireGuard is written in Go, so the network compiles right into a 20 MB binary. Go is made for production servers, LLMs write it well, and the simplicity of Go matches how I like to build. Mainplane works with a Claude or ChatGPT subscription on the regular usage limits, or any API key. I would love your feedback, especially brutal and honest feedback and tasks that the harness cannot do. I'll be here all day to answer questions.
SOMA 是一款开源桌面 CAD 应用：你用自然语言描述需要的零件，Claude Code、Codex CLI、OpenCode 等编程智能体用 Python（CadQuery）写出参数化模型，SOMA 实时构建并显示 3D 结果。适合需要外壳、支架、安装座的创客和 3D 打印爱好者，不会 CAD 也能上手，导出 STEP、STL、3MF 即可打印或加工。 亮点 不绑定任何 AI：使用你自己已有的 CLI 智能体和订阅，模型选择和权限都由你掌控 模型即代码：每个模型都是可读的 CadQuery 脚本，每一步都有 Git 历史，可预览和恢复 和智能体“指着说”：在 3D 视图中点选零件、精确测量、直接在画面上标注，再发送给智能体修改 不只是建模：剖切、爆炸视图、参数面板、标准件（螺钉、螺母、热熔螺母）、FEM 强度分析（CalculiX）、动画渲染和带尺寸的工程图（PDF、DXF） 开箱即用：Windows、macOS、Linux 安装包已内置 Python、CadQuery、Git 和 CalculiX，GPL-3.0 开源 （Note from the maintainer: this text was written with the help of an AI translation, as I don't read Chinese. Replies in English are easiest for me, Chinese is fine too.）
Coding agents write blind. They can't run your app, see the screen, or check their work — so you end up running builds, pasting screenshots, and copying console errors. Vibex is a native Mac window that gives your agent eyes. It embeds a live iOS simulator, browser preview, source control, and AI usage dashboard with 50+ MCP tools. Claude writes code, runs the app, taps the simulator, reads the console, and checks every diff. You just review. Free forever. MIT licensed. No account, no tracking
Manage files side by side on macOS and Windows. Preview files, search contents, explore disk usage, and open a built-in terminal. A keyboard-first alternative to traditional dual-pane file managers for everyday file organization.
RadarQuest is a modern, privacy-respecting search engine designed to cut through SEO spam and cluttered ads. Powered by a multi-source global indexing pipeline and AI Overviews, RadarQuest features dedicated focus lenses for authentic PDFs and exam papers, developer repositories, and unfiltered community discussions.
Give your coding agent local checks and a workflow for shared UI. Find inconsistencies, build and adopt components, and keep examples and decisions with your code. Open source. Works with Codex, Claude Code, and Cursor.
Run Codex, Claude Code, Gemini CLI and more in parallel, each in its own Git worktree. Review every change in one desktop app for Windows, macOS and Linux. Automations, bots, the `jackalope` CLI and Cloudflare Artifacts Swarm mode now in testing and early release.
Talk, and Sorimac types at your cursor in any Mac app: Mail, Slack, Notes, a browser, or a Claude Code prompt in Terminal. Speech recognition runs on your Mac (Qwen3-ASR and Whisper), so it works offline and by default your voice never leaves the Mac. Voice commands, auto-Enter for chat and coding prompts, 21 languages (8 in beta). Pay $9.99 once and use it forever, updates included. 3-day free trial, no account. Apple Silicon, macOS 15+.
Monocurl is a programming language and app for making mathematical animations, videos, and slideshows. It's comparable to Manim, but you see your animation update live as you edit, and you can present it as an interactive slideshow. For students, educators, academics, and anyone who wants to explain an idea visually.
Remote Tools is an all-in-one solution for accessing, monitoring and managing your PCs, servers and IoT devices. Remote desktop, terminal, file transfers, port forwarding and automated tasks - unified in a single high-performance cross-platform application.
AIVerso ranks 138 AI models and tools across 11 categories – chat, coding, image, video, voice, music and more – with benchmark data, prices and free-plan info. Compare up to 3 side by side. Free, no sign-up, no ads.
Build native iOS, Android and web apps with AI and sell them under your own branded platform: your name, logo and domain. Unlimited apps, and our team handles the App Store and Google Play submissions at no additional cost, to both your own and your clients' developer accounts. $85/month, locked in for as long as you stay subscribed.
An advanced client-side AI platform and Web/App No-Code Generator, featuring Bring Your Own Key (BYOK) privacy and P2P video call transfers. Access over 200 free developer tools completely offline with zero data tracking. Try the Offline-First AI Sandbox today! For support, visit DanchoFast.com or contact [email protected]
Lucen builds modular robotics kits for labs, students, and developers. Train bipedal locomotion, then swap six parts to train dexterous manipulation. Programmable platform with onboard GPU. Open firmware and SDK.
I had claude build a course optimized for me (backend engineer with some systems experience) to learn formal methods. Main idea being - I want to build a checked application language which finds bugs with z3, prove things in lean, model concurrency/dist sys behaviors in TLA+, connect it to postgres. It's a static site hosted via gh pages. Sharing primarily for feedback and suggestions from experts as someone who lacks experience in field. My intuition is that what I enjoy about coding and the demand in my field will eventually turn into this sort of thing - where we control agents using proofs and verification systems.
Take notes easily with Notedux, a note-taking tool that uses unique links that serve as both your password and your account, so there's no login. Besides using the browser, you can also create and manage notes directly from the terminal with the open-source Notedux CLI tool.
Bulk-clone and organize your GitHub repos into preset layouts with an AGENTS.md map for AI coding agents. One command via npx on Windows, macOS and Linux.
Hi guys! Built AgentRadar, a Python SDK to track AI agent executions, including tool calls, LLM inferences, token usage, errors, latency and more, all visible on a web dashboard. Integration requires almost zero code changes. Just install it using `pip install agentradar-sdk`, grab your API key, and initialize the SDK. You can enable or disable metadata tracking based on your preferences. It’s lightweight and designed to add almost no latency. Would love your feedback!
When GPT-6/Gemini launched recently, a lot of the 3D demos used blender-mcp ( https://github.com/ahujasid/blender-mcp ). It's cool, but you have to install blender locally and wire up plugins just to try a prompt or share a run. I built a browser-based 3D mesh playground so you can hook up your own coding agent over MCP (or run in-browser) directly in a tab. It runs on three.js + manifold-3d for blender-like boolean/mesh editing in JS, and gives the agent a render_preview tool to take multi-view (front/right/top/iso + custom camera angles + clay mode). That way the agent can look at what it built and fix the geometry, similar to how it iterates when building a website. The submitted link is a trace of Gemini reconstructing a Copenhagen canal scene from a photo: it hits a script error on the first pass, fixes the bug, and then goes through several rounds of checking 4-view and close-up render_preview to tweak the geometry.
Browser console and diagnostic chamber for Oh-My-Pi (omp) agent sessions — live timeline, steering, real PTY, git/diff/wiki panels, and a panel-plugin platform. Elysia + Preact on Bun. - kabmo-labs...
I kept finding myself wanting a way to get talked through the code that a model had created, which led to me making Sóþspell (old English, "Truth Story"). You give it a git locally or via URL, and it creates a tour of the in a web page, with the aim of it explaining the whole thing. There's also model harness plugins for Claude, Codex, and OpenCode. You give it an API key for whatever model you prefer to create the tour (saved locally or via github), and you can also give it a voice using an ElevenLabs API key. I just pushed a canvas view, which I'm hoping I can make the code navigation a bit more helpful. There's a codebase, and you can export tours as JSON. The doesn't need any keys to work, it falls back to browser speech if you want to hear that. This is my first real, "big" open source project that I've released, prior to this I'd just contributed stuff. I hope it can be useful to someone! I would love to hear some constructive criticism too, and any suggestions on features you'd like to see. It's MIT licensed, and I'll try to stick around in the comments to answer questions if people want.
Hey HN! I shared Optio here back in March. Originally it was focused on getting swarms of coding agents (claude code/codex) to drive PRs towards a merge by automatically resuming them on CI failures or PR comments. Since then, I've reworked the project to be a general platform for running any kind of AI workflow on your infra by allowing users to configure each component of a possible workflow: When: On demand, a cron schedule, or triggered by external events such as a webhook, slack message, ticket (jira/linear/gh), pagerduty/datadog/alertsmanager/pylon/sentry. Where: In a k8s pod or any machine you have installed the optio daemon on (I use my laptop). Who: Any harness you prefer (Claude Code, Codex, Gemini, Pi, etc.) What/Why: You can format parameters from the triggers directly into the prompts and then configure if the agent should exit, be resumed automatically, or be persistent. You can either define workflows in the web ui or as yaml files that get synced to the orchestrator (ai config-as-code). I've also built native iOS and Android apps so you can connect to your cluster and create/manage agents remotely. The marketing page is at https://optio.host and the code is still open source at https://github.com/jonwiggins/optio Would love feedback or any thoughts you might have on how I could make it better!
FindAIforWork.com is a curated web directory that catalogs AI software for productive work, sorting 50 reviewed tools into named categories such as Writing & Content, Design & Image, Video & Animation, Audio & Voice, Developer Tools, and Productivity & Meetings.
Tilework runs Claude Code, Codex, Gemini CLI and your shell in real terminal tiles on an infinite canvas. A dot per tile shows who's working, done or waiting on you. Link tiles to hand one agent's result to the next. Free, MIT, Linux.
Build a mini online shop instantly without any monthly SaaS fees, complex coding, or account creation. BytePriva's WhatsApp Store Builder lets freelancers and small business owners generate a clean product catalog and receive direct WhatsApp orders from customers. Best of all, it's 100% browser-based with zero data uploads.
Back in August we found this cool bypass of developer restrictions in Atlantis, leading to an unreviewed terraform apply & RCE! The vulnerable Gitea approval path dates back to the integration's introduction in v0.28.0 (May 2024), so the bypass was most likely present for roughly two and a half years before the patch shipped in October 2026. Enjoy!
cwRadar is an independently developed crypto market intelligence platform monitoring 103 markets. It combines multi-timeframe technical analysis, market structure, supply and demand zones, and risk controls to identify selective trading opportunities. Users receive structured futures signals and position updates via Telegram. Our public Trading Terminal provides access to completed trade histories, including losses. Explore cwRadar with a 7-day free trial.
Inspect any live website like a design file. Explore styles, edit elements in place, audit UX and accessibility, and turn your changes into actionable context for your coding agent. All from a free Chrome extension.
Sell subscriptions with one buy link and keep your app and login. UseThatApp is the merchant of record, handling checkout, sales tax, EU VAT and payouts. For US developers selling to US and EU buyers; no monthly fee. MCP setup for coding agents.
Zavke is a free temporary email service designed for quick, private, and convenient email use without registration. Users can create temporary addresses, receive verification emails, choose custom or random email names, manage inbox history, save favorite messages, and use a temporary address for up to 24 hours. Zavke also offers custom domains, QR access, and a developer API for flexible temporary email workflows.
India-first deploy platform. Git-push hosting for Next.js, Django, Express and FastAPI, with managed MySQL, Mongodb, Postgres Redis(valkey), team workspaces, UPI billing and India data residency.
Hi HN, we're Mike and Thom. Agents of the future should only have one tool (the interpreter, ideally a LISP eval), nothing more or less; to interact with the world. Agents remember by writing markdown files. NODECODE remembers by rewriting itself. And that is in itself a new paradigm in working with agents. What if memories, skills, MCPs, provider connectors, tools, etc. all live under the same image as *.lisp files instead of in disarray across random folders? Ask it for something it can’t do, and it writes what it needs into its OWN running image and keeps it even after a restart. (Changes in a LISP image have significantly lower ceremony and overhead than in harnesses written in other languages): Small intuitive scenario: > you can't read PDFs. teach yourself. (defun read-pdf (path) ...) ; runs at once, kept (read-pdf "invoices/0421.pdf") ;; Globex owes $1,240.00, due Oct 24, unpaid. [… rest of PDF content] read-pdf is still there after a restart, in your terminal, and in every other UI clients (Discord, web app, etc.) Technically at a lower level; NODECODE runs as one process in SBCL, a Common Lisp system, and shells attach to it. The model has one tool: it evaluates LISP in the running program. Code is data, so a program can rewrite itself while it runs. What the model defines runs at once and is kept, one file per definition, committed with git, loaded at every start. Why LISP now? Current SOTA models have reached an optimum point in how they efficiently work with LISP, though not as fluently as Python. Still, we believe AGI will understand ANY language (past/present/future) by understanding the crux of any given language, and not through excessive and expensive pretraining scheme. Why LISP in general? We believe LISP is the current language that communicates to the world with the least bandwidth. Why another LISP agent? The LISP harnesses we found are hard to use unless you already know the language itself, and so we had this idea of simplifying and condensing the UX to be user friendly for the masses, while leveraging the power of LISP. Code as data is what lets an agent change itself, and that shouldn't need a LISP hacker to run. NODECODE is our first step toward putting it on everyone's PC and mobile. You ask in plain words, the model writes LISP, and you get briefed in natural language; that is how it should be. Extensions? Code as data also suits extensions. We see NODECODE as a living agent organism. To us, it feels very much alive, and so this brings us to the Cells system. A Cell is a folder that teaches NODECODE something new: a chat room it can live in (e.g Discord, Telegram, Slack, etc. - Yes we made Cells for channels to have it act as a very capable personal agent), a skill, a memory, MCPs, new tools, etc. Install one from our Cell hub or ask it to write a custom, personalized one. Either way it is more code the agent can read and change. Your terminal, a browser tab, your phone and bots in Discord, Telegram and Slack all attach to that SAME gateway process, so a tool it writes in one is there in ALL of them. Over time it becomes yours and it is continuously evolving; the organism within it; NODECODE. MIT License. https://github.com/nodecode-ai/nodecode X/twitter/demo launch post: https://x.com/pomterree/status/2113721985167728757 docs: https://nodecode.ai/docs
I originally made this addon 4 years ago, but recently discovered that coding agents love it. The addon adds a scripting API for Blender’s node trees, such as Geometry Nodes or Shader Nodes. It generates a function and related types for every node in Blender. Then the agent (or you) can write scripts to create node trees without having to guess what the node, input, and output names are. It’s meant to be used alongside the Blender MCP. In my tests, Opus finishes building node trees in at least half the time using at least half the tokens to answer the same prompt. Because the scripting API forces it to build correct node trees, it has to back track less to fix mistakes. This saves a lot of time and tokens. There is also a skill and some helper functions to let it look up the function signatures for nodes. Let me know how it works for you!
Discover and explore hand-reviewed niche AI tools across business, coding, marketing, and productivity. Ai Niche Tool helps founders, creators, and developers find targeted AI solutions to optimize workflows, increase revenue, and automate daily tasks with ease.
A free alternative to Postman, Insomnia and Bruno that needs no account. REST, GraphQL, gRPC, WebSocket, Socket.IO and MQTT in one app, with collections saved as plain files you can commit. The desktop app, web app and CLI run the same Go engine, so a CI run matches what you saw while testing. Secrets stay in your own manager, and an MCP server lets coding agents send saved requests with values redacted. MIT licensed.
AI chat, coding, web research, images and video in one place. Choose a model and effort for each task, and connect your own tools through a usage-based API.
Hi HackerNews, My name is Matija, and I'm building ranwhat.com. It's used for checking if any of your secrets have leaked, and where they leaked. It's mostly aimed at solo devs such as myself, but also at anyone using a coding agent. The tool also checks if your coding agent ran any sketchy commands. This is mostly a "better safe than sorry" feature, so you can tell your Claude or Codex not to run them. The project is open source (MIT) and everything runs locally on your machine. How to install it: pip install ranwhat / uv tool install ranwhat, or brew install MatijaMiki/ranwhat/ranwhat There is a free and a paid plan. Everything local is free; the paid plan is 12€ a month and adds server-side features, which you can read more about here: https://ranwhat.com/pricing All I'm asking for here is feedback, and if you have anything you would want added to the project. Looking forward to answering any questions you have. Best, Matija
I have built an indie hacker project; my main flow was: 1. chatting with Claud Code about the general plan 2. assigning slices of work to other worker agents like Pi + DeepSeek 4.1 Flash 3. building in parallel 4. Claud Code reviews when done, runs tests, and merges after I validate It worked, but with a lot of glue in the middle, so I built radian + arc into the same binary. Radian: a compact, pragmatic agent that has enough tooling to solve a small coding task end to end; it does not try to be a plugin platform. You can still extend it via MCP, skills, or hooks to extend its capabilities. I think plugins are great, but a lot of them are for humans. I'm trying to remove humans from the loop. Humans are at the start and end of the loop. arc: built into the radian binary; * it can get work sent from another agent * assign work items to concurrent headless radian agents * automatically review, CICD, integrate locally * humans can still be in the loop but not at the agent conversation or code review level, but at the work item approval/review level. * shows all of the above in a plan view or Kanban board in TUI. I have been heavily dogfooding, and Radian + arc basically built themselves in the last couple of weeks, with me just being a product manager/scrum master and qa. radian + arc are open source at https://github.com/hisea/radian What do you think about this approach? Looking for ideas to make Radian better or fit more scenarios.
Hi HN, My name is Johny Georges and I would like to present Jet, a C-like general purpose programming language. It's currently in alpha so I wouldn't recommend using it in a professional environment. Here are some quick facts about it: - it supports both OOP and functional programming paradigms - the syntax is similar to Typescript and Swift - arrays and strings are dynamic - memory is automatically managed for you - you have full control over how values get passed around (by value or by reference) In anticipation of the typical questions that come with creating and publishing a new programming language, I've authored an FAQ that's available in the repo. I will also be around today to answer any other questions that come up. FAQ: https://github.com/JetJet13/Jet/blob/main/FAQ.md Special thanks to Robert Nystrom for authoring 'Crafting Interpreters' and making it available for free, what a guy! I also want to give props to Chris Lattner & Vikram Advethe for creating LLVM. What an impressive piece of kit!
ARDA is an investment real estate aggregator for Bali. Developers update prices, unit availability and payment plans themselves, so agents and investors always see current data. Search by map and filters, open a project card with units, prices, permits and financial models, build a commercial offer in a few clicks and message the developer directly. We don't sell property and don't take commission. Free for agents and buyers.
ANKITA turns your Windows desktop into an AI workspace. Delegate to subagents and follow their progress, tool calls and results. Work with files and code, control the browser and desktop, connect MCP tools and plugins, and keep projects, memory and scheduled tasks together. Use your ChatGPT account or choose another provider. Includes a floating companion, reusable skills and a terminal CLI. Free and open source; provider costs and limits apply.
Cursor only offers light & dark mode, but I've grown accustomed to Nord's low-contrast, legible palette (great for long sessions). So I wrote a Nord glass theme into its theme store: chrome, git colors, terminal ANSI.
AI Bridge runs Claude Code, Codex, Grok and OpenCode side by side on your Mac: each the real CLI in a real terminal, on your own subscription. Group your agents like a team, by department or by what each one owns. Open a Kubernetes cluster or talk to your database in plain words, right beside them. Prefer terminals? The Grid puts them all on one wall, with cluster and database chats where you need them. Open as many windows as you like, and put agents in Slack that reply when you approve.
Catchenv secrets leaking into git: a pre-commit hook and CLI that fails when an API key or any value of your env files is staged, tracked or anywhere in the history. Your real values, not patterns.
An orchestrator splits your work across Claude Code and Codex agents, each in its own git worktree. A reviewer checks the result, a built-in browser lets agents test their own work, and every question lands in one place. Uses the plans you already have. Free for 14 days.
Tired of bloated new-tab pages tracking your every move? Hacker New Tab is a lightning-fast, keyboard-driven retro terminal dashboard with 4 distinct modes (Personal, Research, Dev, CTF), instant IOC pivots, inline decoders, per-mode scratchpads, zero ads, and zero telemetry. All local.
A terminal multiplexer built for people running several coding agents at once. Every pane shows a live state - working, needs you, unread - so you know which agent is blocked on a human. Like tmux, agents and dev servers keep running when you detach. Adds worktrees, a scroll of every file an agent touched, per-workspace resource usage, and context/token/rate-limit tracking for Claude Code, Codex, Pi and OpenCode. macOS and Linux (Apple Silicon/amd64), one curl install.
Hey HN! As an avid user of cmux ( https://cmux.com/ ), I wanted a way to control terminal and agent sessions from my Android phone, so I built Muxer. The app talks to a tiny relay over HTTPS, that then talks to cmux through its local socket, and gives you to ability to: - Answer agent permission requests, questions and plans from a notification - Reply to Claude Code, Codex and Pi chat sessions with text, voice, @ file mentions and photos - Receive all cmux notification, with inline replies, bubbles and background delivery. - Use a live terminal with sticky modifier keys and local scrollback - Share text and links from any app to an agent - Organize workspaces - Manage cmux Cloud VMs and automations - Lock the app with your fingerprint, face or screen lock Enjoy, and let me know what you think!
Most ADEs bring the browser into the IDE. tabsh takes the opposite approach: bring coding sessions to your browser, where your work already happens. It's a small Rust daemon that puts all your sessions across projects on one Kanban board/list, with status updates via agent hooks. Use real terminals or chat panels with Claude Code, Codex, Cursor CLI, OpenCode, and more. Free, MIT-licensed, macOS/Linux.
TamaLAUNCHER is a minimalist, frictionless Android launcher designed for speed and simplicity. Keep your existing home screen setup while accessing your favorite apps instantly from a clean launch cockpit. Key Features: - Frictionless UI: Visual [➕] -> [⭐] -> [🚀] workflow. - No Clutter: Access apps without messing up your home screen. - Fast & Lightweight: Zero bloat, pure launching speed. Built by a solo developer for a smooth daily workflow.
Coding agents rewrite code you never asked them to touch, and the tests still pass. Sonah reads the agent’s Claude Code or Codex session along with the code, then shows on one page: whether each ask was done, what the agent decided on its own, which of its claims its own results contradict, and what is wrong in the code. Every quote is checked word for word. Runs locally through your own Claude login. No account, no telemetry. Free for individuals.
Hand off work between coding agents: Claude Code to Codex, Codex to Devin, and more. Every session kept on your machine, searchable and replayable. Open source, local-first.
78 free tools in one place: India tax and salary calculators, unit converters, text and developer utilities, and PDF tools. Everything runs in your browser, so files you drop in never leave your device. No account, no limits. Every calculator shows its formula.
**Dating beyond the swipe 💜** Tired of endless swiping and conversations that go nowhere? I'm building **MeetSoulWeave**, a dating platform focused on helping people explore more meaningful connections. I'm looking for people in the US who are willing to try something new and share honest feedback. What do you think today's dating apps are missing? Discover MeetSoulWeave: I'm the developer behind the project, and I'd love to hear your thoughts!
Since our first launch: 22 updates in 17 days, many shipped the same day a user asked. SwarmPane runs Claude Code, Codex, Gemini, Cursor and 9 more AI coding agents side by side, on the plans you already pay for. New: every account's usage limits in your toolbar, each agent in its own worktree, Devin and Hermes, one rules file and MCP setup for every agent, and sessions that survive restarts. Mac, Windows, Linux.
MiniMax M3.1 Flash is MiniMax’s new Flash model for agentic tasks, coding, computer use and cybersecurity. MiniMax says it is about 3x faster online and 70% cheaper for agentic tasks than its predecessor, with output at 200+ tokens per second. Pricing is $0.15 input and $0.60 output per 1M tokens. It’s on the MiniMax API, M Plan and MiniMax Code, and weights are promised later.
Turn your ideas into real retro games. Describe, play, refine, and download your game—no coding needed. Create for Game Boy, Game Boy Color, Game Boy Advance, NES, SNES, Famicom, and Super Famicom. Get a physical cartridge release and let Retroable publish your game for you. Or discover and play community-created games for free, directly in your browser. Your idea. Your game. Real hardware.
Hi HN! I made git-dupe to solve the problem where I have notes, skills, and agent instructions that are part of my personal workflow, so I do not want them committed to the public repo, but at the same time, I want them to be in the project at the correct locations (e.g., AGENTS.md file locations matter), and these files cannot be simply gitignored because I want versioning for them. The solution that I've come upon is versioning those agent files in another repo, say <private-repo> (this is private to me and holds the history for my agent files). The project itself is open to the public in another <public-repo>. I overlap the two repos to solve the problem (i.e., they share the same worktree). $ git status --short # the public ?? AGENTS.md ?? notes/ $ git dupe init # the private Initialized empty Git repository in /home/brian/code/project/.git/dupe/ $ git dupe add notes/ AGENTS.md # the private hint: notes is now hidden from the project's Git; run from the root, 'git dupe unhide -- notes' stops hiding it $ git dupe commit -m "Notes and agent instructions" # the private [main (root-commit) 3eb3af9] Notes and agent instructions 3 files changed, 3 insertions(+) create mode 100644 .gitdupe create mode 100644 AGENTS.md create mode 100644 notes/todo.md $ git status --short # the public nothing $ git dupe log --oneline # the private your files 3eb3af9 Notes and agent instructions In short, git-dupe allows two Git repos to share the same worktree, while using native Git for most operations. Most git-dupe commands behave just like normal Git commands, operating on the private instead. git-dupe currently supports Linux and macOS, and is under the MIT License; the git-dupe contains much more extensive technical documentation documentation for those who are interested. If you have any feedback/questions about git-dupe, feel free to message. I hope it helps! Fun fact: the git-dupe project itself is using git-dupe for versioning.
VibeGuard is an AST-based security auditing tool designed specifically for AI-generated code. As developers increasingly rely on AI coding assistants (like Cursor, Copilot, and Claude), maintaining code quality and security becomes a major challenge. VibeGuard scans your codebase locally using advanced Abstract Syntax Tree analysis with zero code retention, ensuring your IP and sensitive data never leave your machine while catching vulnerabilities, secret leaks, and bad patterns before you ship.
An AI-powered coding agent built for developers who want to work directly from the terminal. Use supported AI providers, connect with an API key, and bring AI-assisted coding into your development workflow.
Orcus is the AI SDK. With Orcus, No LLM API key required to make AI model requests. Orcus uses your web credentials to access AI models available through supported provider websites and deliver responses to your applications through a familiar, API-style interface. From model invocation and streaming to structured outputs, tool calling, and batch processing, Orcus brings essential AI integration capabilities into one developer-friendly SDK.
