🛠️ Hacker News Tools

[API/SDK] Show HN: SwiftAI – open-source library to easily build LLM features on iOS/macOS We built SwiftAI, an open-source Swift library that lets you use Apple’s on-device LLMs when available (Apple opened access in June), and fall back to a cloud model when they aren’t available — all without duplicating code.<p>SwiftAI gives you: - A single, model-agnostic API - An agent&#x2F;tool loop - Strongly-typed structured outputs - Optional chat state<p>Backstory: We started experimenting with Apple’s local models because they’re free (no API calls), private, and work offline. The problem: not all devices support them (older iPhones, Apple Intelligence disabled, low battery, etc.). That meant writing two codepaths — one for local, one for cloud — and scattering branching logic across the app. SwiftAI centralizes that decision. Your feature code stays the same whether you’re on-device or cloud.<p>Example<p><pre><code> import SwiftAI let llm: any LLM = SystemLLM.ifAvailable ?? OpenaiLLM(model: &quot;gpt-5-mini&quot;, apiKey: &quot;&lt;key&gt;&quot;) let response = try await llm.reply(to: &quot;Write a haiku about Hacker News&quot;) print(response.content) </code></pre> It&#x27;s open source — we&#x27;d love for you to try it, break it, and help shape the roadmap. Join our discord &#x2F; slack or email us at root@mit12.dev.<p>Links<p>- GitHub (source, docs): <a href="https:&#x2F;&#x2F;github.com&#x2F;mi12labs&#x2F;SwiftAI" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;mi12labs&#x2F;SwiftAI</a><p>- System Design: <a href="https:&#x2F;&#x2F;github.com&#x2F;mi12labs&#x2F;SwiftAI&#x2F;blob&#x2F;main&#x2F;Docs&#x2F;Proposals&#x2F;001-llm-api.md" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;mi12labs&#x2F;SwiftAI&#x2F;blob&#x2F;main&#x2F;Docs&#x2F;Proposals...</a><p>- Swift Package Index (compat&#x2F;builds): <a href="https:&#x2F;&#x2F;swiftpackageindex.com&#x2F;mi12labs&#x2F;SwiftAI" rel="nofollow">https:&#x2F;&#x2F;swiftpackageindex.com&#x2F;mi12labs&#x2F;SwiftAI</a><p>- Discord <a href="https:&#x2F;&#x2F;discord.com&#x2F;invite&#x2F;ckfVGE5r" rel="nofollow">https:&#x2F;&#x2F;discord.com&#x2F;invite&#x2F;ckfVGE5r</a> and slack <a href="https:&#x2F;&#x2F;mi12swiftai.slack.com&#x2F;join&#x2F;shared_invite&#x2F;zt-3c3lr6dat-jJ8BHBsdWc47o4FDu2CgHQ#&#x2F;shared-invite&#x2F;email" rel="nofollow">https:&#x2F;&#x2F;mi12swiftai.slack.com&#x2F;join&#x2F;shared_invite&#x2F;zt-3c3lr6da...</a>

[Other] Show HN: GrowChief – open-source social media outreach tool

[IDE/Editor] Show HN: AI Agent in Jupyter – Runcell I build runcell, an AI Agent in Jupyter Lab. It can understand context (data, charts, code, etc) in your jupyterlab and write code for you.<p>Runcell has built-in tools that can edit or execute cells, read&#x2F;write files, search web, etc.<p>Comparing with AI IDE like cursor, runcell focus on building context for code agent in jupyter environment, which means the agent can understand different types of information in jupyter notebook, access kernel state, edit&#x2F;execute specific cells instead of handling jupyter as static ipynb file.<p>Comparing with jupyter ai, runcell is more like an agent instead of a chatbot. It have access to lots of tools to work and take actions by its own.<p>You can use runcell with simple &quot;pip install runcell&quot; to start.<p>Any comments and suggestions are welcome.

[Other] Show HN: I built AI that turns 4 hours of financial analysis into 30 seconds I built Duebase AI to solve a problem I kept running into in fintech - analyzing UK company financial health takes forever. The process usually goes: download PDFs from Companies House → manually extract data to spreadsheets → calculate ratios → interpret trends. Takes 3-4 hours per company and requires serious financial expertise. The technical challenge: Companies House filings are messy. Inconsistent formats, complex accounting structures, missing data, and you need to understand UK accounting standards to make sense of it all. My approach:<p>Parse 15M+ UK company records from Companies House API Built ML models to extract and normalize financial data from varied filing formats Created scoring algorithms that weight liquidity, profitability, leverage, and growth trends Generate 1-5 health scores with explanations in plain English<p>What it does:<p>Instant financial analysis of any UK company (30 seconds vs 4 hours) Real-time monitoring with alerts for new filings&#x2F;director changes Risk detection that catches declining trends early No financial background needed to understand results<p>The hardest part was handling the data inconsistencies - UK companies file in different formats, use various accounting frameworks, and often have incomplete information. Had to build a lot of data cleaning and normalization logic. Currently focused on the UK market since I know the regulatory landscape well, but the approach could work for other countries with similar public filing systems. Link: <a href="https:&#x2F;&#x2F;duebase.com" rel="nofollow">https:&#x2F;&#x2F;duebase.com</a>

[Other] Show HN: Meetup.com and eventribe alternative to small groups Mobile first open-source RSVP platform. Alternative for meetup.com &#x2F; eventribe for small companies and groups. If you have a small group and don&#x27;t want to pay for services you can easily selfhost this solution. Open for improvements and for feedback, ofc.<p>- One-Click Sharing - Each event gets a unique, memorable URL. Share instantly via any platform or messaging app. - No Hassle, No Sign-Ups - Skip registrations and endless forms. Unlike other event platforms, you create and share instantly — no accounts, no barriers. - Effortless Simplicity - Designed to be instantly clear and easy. No learning curve — just open, create, and go.

[Other] Show HN: React Web Camera – Fix <input type=file> single-photo limit What we built<p>React Web Camera is a lightweight, reusable React component that allows users to capture multiple photos in one camera session, in-browser. It works across standard web apps, responsive UIs, and Progressive Web Apps (PWAs)—unlocking a smoother experience than the default &lt;input type=&quot;file&quot; capture&gt; element.<p>The problem<p>On mobile (and increasingly on desktops), using: &lt;input type=&quot;file&quot; accept=&quot;image&#x2F;*&quot; capture=&quot;environment&quot;&gt; only allows taking one picture before the camera closes. Want to add more? You have to reopen it each time.<p>How React Web Camera solves it<p>Opens the camera inline in-browser, Lets the user capture multiple photos in one go, Allows previewing captured photos, removing unwanted ones, and submitting everything in a batch, Fully client-side, respects user privacy, Supported across web, responsive UIs, and installable PWAs.

[Other] Show HN: I made an Animal Crossing style letter editor I made a simple open-source letter editor inspired by Animal Crossing NH. Took me forever to look over each card, but I&#x27;m quite pleased with how it turned out. You can even click the bottle in the bottom right to see a random letter design shared by other users! Now to see how long it stays up...<p>Check out the source code here: <a href="https:&#x2F;&#x2F;github.com&#x2F;IdreesInc&#x2F;Animal-Crossing-Letter-Generator" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;IdreesInc&#x2F;Animal-Crossing-Letter-Generato...</a>

[CLI Tool] Bring Your Own Agent to Zed – Featuring Gemini CLI <a href="https:&#x2F;&#x2F;developers.googleblog.com&#x2F;pt-br&#x2F;gemini-cli-is-now-integrated-into-zed&#x2F;" rel="nofollow">https:&#x2F;&#x2F;developers.googleblog.com&#x2F;pt-br&#x2F;gemini-cli-is-now-in...</a>

[Other] Running our Docker registry on-prem with Harbor

[Other] ASCIIFlow

[Other] Show HN: Auto-Match – How We Built Receipt-to-Transaction Matching (Open Source) I’ve been working on automating bookkeeping tasks, and one big pain point was manually reconciling receipts with bank transactions. We built a system that runs in the background, parses receipts (including Gmail), suggests matches, and learns from confirmations to auto-match over time.<p>It&#x27;s built into Midday and fully open-source.<p>Let me know if you have any questions!

[Other] Show HN: Envoy – Command Logger Envoy is a lightweight, background utility that logs your terminal commands. It&#x27;s designed to be a simple and unobtrusive way to keep a history of your shell usage, which can be useful for debugging, tracking work, or just remembering what you did.

[Other] Terminal sessions you can bookmark

[Other] Show HN: Regolith – Regex library that prevents ReDoS CVEs in TypeScript I wanted a safer alternative to RegExp for TypeScript that uses a linear-time engine, so I built Regolith.<p>Why: Many CVEs happen because TypeScript libraries are vulnerable to Regular Expression Denial of Service attacks. I learned about this problem while doing undergraduate research and found that languages like Rust have built-in protection but languages like JavaScript, TypeScript, and Python do not. This library attempts to mitigate these vulnerabilities for TypeScript and JavaScript.<p>How: Regolith uses Rust&#x27;s Regex library under the hood to prevent ReDoS attacks. The Rust Regex library implements a linear-time Regex engine that guarantees linear complexity for execution. A ReDoS attack occurs when a malicious input is provided that causes a normal Regex engine to check for a matching string in too many overlapping configurations. This causes the engine to take an extremely long time to compute the Regex, which could cause latency or downtime for a service. By designing the engine to take at most a linear amount of time, we can prevent these attacks at the library level and have software inherit these safety properties.<p>I&#x27;m really fascinated by making programming languages safer and I would love to hear any feedback on how to improve this project. I&#x27;ll try to answer all questions posted in the comments.<p>Thanks! - Jake Roggenbuck

[Other] GNU Artanis – A fast web application framework for Scheme

[IDE/Editor] Show HN: First background agents in Jetbrains IDEs [video] TLDR: made the first background coding agent that has an isolated workspace and runs locally<p>Howdy - I’m Kevin, co-founder of Firebender, and we built the first background coding agent in android studio &#x2F; Jetbrains!<p>Why not just use Cursor background agents or OpenAI Codex?<p>Both of these require setting up a cloud container and cloning your existing developer environment, and maintaining it. Then when you want to iterate on changes as AI inevitably makes a mistake, you either throw away the work, or have to pull down the branch and clean it up. This feels really clunky. With Firebender, background agents run locally in a lightweight git worktree&#x2F;IDE tab. This means when the agent is done, you can easily clean up the changes and run code with a few clicks.<p>Under the hood, the agent behaves similarly to claude code (didn’t want to reinvent the wheel), but also leverages all of the hooks into IntelliJ sdk like go-to-definition, find usages, auto-imports for accuracy, and it gives a cleaner visual UI for reviewing changes and merging them. You can use any frontier model like gpt-5&#x2F;sonnet-4 as the base.<p>We’ve had to do quite a bit of reverse engineering of the IntelliJ codebase to cleanly set up and manage the isolated environment, and I think you’ll appreciate the simple UX of hitting cmd+enter to run a background agent anywhere.<p>get started docs: <a href="https:&#x2F;&#x2F;docs.firebender.com&#x2F;get-started&#x2F;background-agents">https:&#x2F;&#x2F;docs.firebender.com&#x2F;get-started&#x2F;background-agents</a><p>download the plugin: <a href="https:&#x2F;&#x2F;firebender.com">https:&#x2F;&#x2F;firebender.com</a><p>Would love to get your feedback to help us improve the tool for you! Thanks!

[Code Quality] Show HN: Framework to create linters for Python, YAML, TOML, JSON Hi, this is a Python framework I have created to make my life a little easier when creating custom linting rules.<p>Key features:<p>- Quick to use (~20 LOC to create a functioning linter&#x2F;checker)<p>- Works with different file formats (as in the description)<p>- Supports multiple ignore&#x2F;noqa directives (per-line ignores, range ignores, whole file, rules choice during runs)<p>- Single check can span all elements of the file or even all files<p>- Works directly with Python&#x27;s AST<p>- Tutorials and API reference to make users life easier<p>All in all seems to be quite functional for my private (yet) linters, so I decided to make it into a FOSS project.<p>Have fun, would love to hear your feedback and ideas regarding it!

[Other] Show HN: My OSS P2P file transfer tool for learning Next.js (as a C++ dev) Hey HN,<p>I&#x27;m a C++ algorithms engineer, and today I&#x27;d like to share my first full-stack web project: PrivyDrop.<p>This project was born from two ideas:<p>First, I wanted to solve a daily annoyance of mine: I needed a simple, AirDrop-like way to send text, links, or screenshots between my phone and PC. I tried many tools, but they either required me to sign up, had various limitations, or uploaded my data to their servers, which I was never comfortable with.<p>Second, this was a personal experiment. Last year, with all the talk about &quot;AI replacing programmers,&quot; I got curious. As a developer with no web background, I wanted to see how long it would take to learn full-stack development from scratch (an area I&#x27;m really interested in) and build a complete application, using AI as my primary coding partner and mentor.<p>PrivyDrop is the result of that learning and experiment.<p>It&#x27;s a peer-to-peer (P2P) file-sharing tool based on WebRTC. Its core principle is that *your data belongs only to you*: all files and text are transferred directly between your browser and the recipient&#x27;s, fully end-to-end encrypted, without ever passing through an intermediate server.<p>*Here are some of its key features:*<p>* Completely free and open-source, with no ads. * No sign-up or installation required—just open your browser. * Direct P2P connection for privacy, security, and speed. * No limits on file size or type. * Support for entire folder transfers. * Built-in resumable transfers.<p>The whole process has been a fantastic learning journey. I&#x27;m sharing it now in the hope that it can solve the same pain point for others. I&#x27;m really eager to hear any feedback, ideas, or even harsh criticism from the HN community!<p>* *Live Demo:* <a href="https:&#x2F;&#x2F;www.privydrop.app" rel="nofollow">https:&#x2F;&#x2F;www.privydrop.app</a> * *GitHub Repo:* <a href="https:&#x2F;&#x2F;github.com&#x2F;david-bai00&#x2F;PrivyDrop" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;david-bai00&#x2F;PrivyDrop</a><p>Thanks, everyone!

[Other] Show HN: Rebuilding GPT2 inference in ~500 lines of (commented) code

[API/SDK] Show HN: Mixing Deterministic Codegen with LLM Codegen for Client SDKs Hi HN, I’m Patrick. Elias, Kevin, and I are building Sideko (<a href="https:&#x2F;&#x2F;sideko.dev" rel="nofollow">https:&#x2F;&#x2F;sideko.dev</a>), a new type of code generator for building and maintaining API client SDKs from OpenAPI specs.<p>Our approach differs significantly from traditional SDK generators in that we use structured pattern matching queries to create and update the code. Other SDK generators use templates, which overwrite custom changes and produce code that looks machine generated.<p>We’ve mixed in LLM codegen by creating this workflow: Run deterministic codegen to establish the SDK structure. Let LLMs enhance specific components where adaptability adds value and include agent rules files that enforce consistency and correctness with type checking and integration tests against mock servers. The system will retain the LLM edits, while the rest of the SDK is automatically maintained by the deterministic generator (keeping it in sync with the API). LLMs can edit most the files (see python rules and typescript rules).<p>You can try it out from your terminal: Install: npm install -g @sideko&#x2F;cli Login: sideko login Initialize: sideko sdk init Prompt: “Add a new function that…”<p>Check out the repo for more details: <a href="https:&#x2F;&#x2F;github.com&#x2F;Sideko-Inc&#x2F;sideko" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Sideko-Inc&#x2F;sideko</a> We’d love to hear your thoughts!