🛠️ All DevTools

[Other] Structured Outputs on the Claude Developer Platform (API)

[Other] Show HN: spymux – Spy on your tmux panes I had motivation for writing this after I kept switching back and forth between agents to see if they&#x27;ve finished what they were working on (and I couldn&#x27;t find a similar tool out there). I&#x27;d imagine it can be useful for other scenarios as well, e.g. tracking multiple build&#x2F;test jobs over multiple windows.<p>Still a work in progress, but I thought I&#x27;d share :)

[CLI Tool] Show HN: TalkiTo – enabling voice and Slack for Claude Code and Codex CLI Hey everyone, here is an open source project I&#x27;ve been working on to add voice input&#x2F;output to terminal based coding agents.<p>One thing about the new terminal coding agents I really like is being able to multi-task but right now it&#x27;s a bit like a Tesla on autopilot needing your hands still on the wheel. You need to be checking often if your input is required or if it&#x27;s going off the rails. To be able to go fully hands free I wanted to add TTS and ASR. Then I added slack and WhatsApp hooks to TalkiTo as well.<p>It&#x27;s fully open source with a BYOK philosophy and it&#x27;s configured to work with any of the major ASR&#x2F;TTS providers. It also supports local whisper and kokoro&#x2F;kittentts if you want a decent free&#x2F;private option.<p>It works by wrapping the coding agent and capturing the input&#x2F;output. It does have an MCP server running but thats mainly for configuration - I found that using MCP to speak or listen was too slow. The upshot of the MCP server is you can type (or say) &quot;talkito disable ASR&quot; or &quot;talkito change tts to kokoro&quot;.<p>Here is a demo video I made here: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=pf8jFt0smqs" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=pf8jFt0smqs</a><p>I like to think of it as similar to SuperWhisper but with TTS, the focus on coding agents and configurability.<p>Really curious to get feedback.<p>Thanks!

[Other] Magit manuals are available online again

[Other] Show HN: European tech news in 6 languages I built a multilingual news aggregator for European tech in 6 languages (EN, DE, FR, ES, IT, NL), filterable by audience (consumers, businesses, government) to help users discover relevant European alternatives.<p>A few interesting technical challenges:<p>PATTERN-BASED IMAGE GENERATION Generic AI images all look the same (&quot;holographic businessman with laptop&quot;). I solved this with a JSON database of 60+ concrete visual patterns—context-aware selection prevents repetition. Funding news → coins&#x2F;contracts, security → locks&#x2F;shields. Much more visual diversity. Still needs tweaking though.<p>GRADUAL SITEMAP GROWTH FOR NEW SITES Daily news pipeline generates lots of content. Challenge: Google indexes new sites slowly—exposing all 2K+ URLs at once resulted in 1-3% indexing rate. Solution: Dynamic rolling window in sitemap—only expose top 50 most recent articles (300 URLs), gradually increasing to 100 → 150 → 200 over 6 months as site builds authority. Result: Indexing rate improved to 10%.<p>AUTOMATED TRANSLATION PIPELINE RSS aggregation → AI summaries and translation with context-aware prompts → human review queue. Still tweaking quality per language pair.<p>STACK: Next.js 15 with Turbopack, PostgreSQL, deployed via Sliplane@Hetzner which works pretty ok.<p>STATUS: 1+ month post launch, 80 pages indexed, 100+ European apps featured (Mollie, Wire, Ecosia, LanguageTool, etc.)<p>FEEDBACK WELCOME ON: • European news sources I&#x27;m missing? • Better approaches to multilingual SEO? • Translation quality red flags? • How to optimize for LLM discovery (ChatGPT, Claude, etc.)?

[DevOps] Show HN: Encore – Type-safe back end framework that generates infra from code

[Other] Reproducible C++ builds by logging Git hashes

[CLI Tool] CLI tool to check the Git status of multiple projects

[DevOps] AI-compatible Linux server management for macOS AI-compatible Linux server management app for macOS. Control your infrastructure through AI agents with integrated MCP server, or use traditional tools. Native Mac app with multitab terminals, real-time monitoring, SSH tunneling, and system administration.

[Other] Track incidents Incident tracking and public status pages for small teams. IncidentPulse is a self-hosted, open-source web app that helps developers log incidents, assign teammates, and publish real-time status updates—all from one clean dashboard.

[Testing] RegreSQL: Regression Testing for PostgreSQL Queries

[Database] Show HN: DBOS Java – Postgres-Backed Durable Workflows Hi HN - I’m Peter, here with Harry (devhawk), and we’re building DBOS Java, an open-source Java library for durable workflows, backed by Postgres.<p><a href="https:&#x2F;&#x2F;github.com&#x2F;dbos-inc&#x2F;dbos-transact-java" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;dbos-inc&#x2F;dbos-transact-java</a><p>Essentially, DBOS helps you write long-lived, reliable code that can survive failures, restarts, and crashes without losing state or duplicating work. As your workflows run, it checkpoints each step they take in a Postgres database. When a process stops (fails, restarts, or crashes), your program can recover from those checkpoints to restore its exact state and continue from where it left off, as if nothing happened.<p>In practice, this makes it easier to build reliable systems for use cases like AI agents, payments, data synchronization, or anything that takes hours, days, or weeks to complete. Rather than bolting on ad-hoc retry logic and database checkpoints, durable workflows give you one consistent model for ensuring your programs can recover from any failure from exactly where they left off.<p>This library contains all you need to add durable workflows to your program: there&#x27;s no separate service or orchestrator or any external dependencies except Postgres. Because it&#x27;s just a library, you can incrementally add it to your projects, and it works out of the box with frameworks like Spring. And because it&#x27;s built on Postgres, it natively supports all the tooling you&#x27;re familiar with (backups, GUIs, CLI tools) and works with any Postgres provider.<p>If you want to try it out, check out the quickstart:<p><a href="https:&#x2F;&#x2F;docs.dbos.dev&#x2F;quickstart?language=java" rel="nofollow">https:&#x2F;&#x2F;docs.dbos.dev&#x2F;quickstart?language=java</a><p>We&#x27;d love to hear what you think! We’ll be in the comments for the rest of the day to answer any questions.

[Other] Show HN: Agent-to-code JIT compiler for Z3-theorem-proving agents

[IDE/Editor] Cursor, but open-source and not spying on you. CortexIDE is an open source Cursor alternative. Full privacy. Fully-featured.

[Build/Deploy] Deploy Static Sites in Seconds PinMe is a lightweight CLI tool for deploying static sites to decentralized storage. Permanent, simple, and fast. Perfect for developers who want to publish instantly and keep their content forever.

[API/SDK] Customize powerful AI agents in seconds. Prompt Lyfe helps you customize and build AI agents without writing code. Define your agent’s role, experience, and task, and Prompt Lyfe generates a structured developer prompt ready for ChatGPT, Claude, and more. No prompt engineering skills required. Built by a Data Scientist specializing in AI, it also includes a unique traceability feature that explains why each prompt will be effective, helping users understand the logic behind every AI instruction.

[Other] Show HN: I made an open-source Rust program for memory-efficient genomics My cofounder and I run a startup in oncology, where we handle cancer genomics data. It occurred to me that, thanks to a recent complexity theory result, there&#x27;s a clever way to run bioinformatics algorithms using far less RAM. I built this Rust engine for running whole-genome workloads in under 100MB of RAM. Runtime is a little longer as a result - O(TlogT) instead of O(T). But it should enable whole-genome analytics on consumer-grade hardware.

[CLI Tool] Show HN: ChatExport Structurer – parse ChatGPT/Claude exports into queryable SQL I wanted to query my own chat history but the JSON exports were a mess to work with. Built a small parser that turns them into clean SQL databases. Parsed 70k+ messages across multiple models. Useful for analyzing chat history, building a personal knowledge base, or archiving conversations. Simple CLI, open source.

[Other] Run Windows Subsystem For Android on your Windows 10 and Windows 11 PC using prebuilt binaries with Google Play Store (MindTheGapps) and/or Magisk or KernelSU (root solutions) built in.

[Testing] Show HN: Tokenflood – simulate arbitrary loads on instruction-tuned LLMs Hi everyone, I just released an open source load testing tool for LLMs:<p><a href="https:&#x2F;&#x2F;github.com&#x2F;twerkmeister&#x2F;tokenflood" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;twerkmeister&#x2F;tokenflood</a><p>=== <i>What is it and what problems does it solve?</i> ===<p>Tokenflood is a load testing tool for instruction-tuned LLMs hat can simulate arbitrary LLM loads in terms of prompt, prefix, and output lengths and requests per second. Instead of first collecting prompt data for different load types, you can configure the desired parameters for your load test and you are good to go. It also let&#x27;s you assess the latency effects of potential prompt parameter changes before spending the time and effort to implement them.<p>I believe it&#x27;s really useful for developing latency sensitive LLM applications and * load testing self-hosted LLM model setups * Assessing the latency benefit of changes to prompt parameters before implementing those changes * Assessing latency and intraday variation of latency on hosted LLM services before sending your traffic there<p>=== <i>Why did I built it?</i> ===<p>Over the course of the past year, part of my work has been helping my clients to meet their latency, throughput and cost targets for LLMs (PTUs, anyone? ). That process involved making numerous choices about cloud providers, hardware, inference software, models, configurations and prompt changes. During that time I found myself doing similar tests over and over with a collection of adhoc scripts. I finally had some time on my hands and wanted to properly put it together in one tool.<p>=== <i>What am I looking for?</i> ===<p>I am sharing this for three reasons: Hoping this can make other&#x27;s work for latency-sensitive LLM applications simpler, learning and improving from feedback, and finding new projects to work on.<p>So please check it out on github (<a href="https:&#x2F;&#x2F;github.com&#x2F;twerkmeister&#x2F;tokenflood" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;twerkmeister&#x2F;tokenflood</a>), comment, and reach out at thomas@werkmeister.me or on linkedin(<a href="https:&#x2F;&#x2F;www.linkedin.com&#x2F;in&#x2F;twerkmeister&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.linkedin.com&#x2F;in&#x2F;twerkmeister&#x2F;</a>) for professional inquiries.<p>=== Pics ===<p>image of cli interface: <a href="https:&#x2F;&#x2F;github.com&#x2F;twerkmeister&#x2F;tokenflood&#x2F;blob&#x2F;main&#x2F;images&#x2F;cli.png?raw=true" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;twerkmeister&#x2F;tokenflood&#x2F;blob&#x2F;main&#x2F;images&#x2F;...</a><p>result image: <a href="https:&#x2F;&#x2F;github.com&#x2F;twerkmeister&#x2F;tokenflood&#x2F;blob&#x2F;main&#x2F;images&#x2F;self-hosted_shorter_output_latency_percentiles.png?raw=true" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;twerkmeister&#x2F;tokenflood&#x2F;blob&#x2F;main&#x2F;images&#x2F;...</a>