🛠️ Hacker News Tools

[Other] ChunkLLM: A Lightweight Pluggable Framework for Accelerating LLMs Inference

[Other] Show HN: Inspec – Specification scheduling software for interior designers Hi HN, I&#x27;m Nick and I built Inspec with my partner Caitlin (she&#x27;s the interior designer, I&#x27;m the developer).<p>Inspec lets interior designers create and manage FF&amp;E schedules - basically spec sheets listing everything that goes in a room before the furniture: flooring, paint, countertops, lighting, fixtures, etc.<p>This process is currently exclusively done using Excel, and while this works to some degree, it comes with a lot of manual work and clunkiness. When Caitlin started working remotely, I watched her deal with these issues daily. So I spent the past 9 months building a better solution.<p>Key Features:<p>- Realtime collaboration<p>- Revision control (versioning)<p>- Professional PDF exports<p>- QR codes to print out on-site for contractors, builders etc. to always have the latest version<p>- Completely customisable fields and a familiar workflow to Excel to keep learning curve small<p>When we first started building this we thought it was completely unique, but turns out there are 2 competitors. I think our positioning is quite different though as we seek to only replace this document creation, rather than overhaul the entire project management and try to be a complete solution. So that’s helped to guide our focus, but also validated that interior designers are looking to modernise their software.<p>Tech stack:<p>I used my SaaS boilerplate which is just a modified T3 stack template with everything set up just how I like it. So Next.js with the pages router, TS, Trpc, Prisma + PostgreSQL, and TailwindCSS. I’ve also got Pusher for realtime collaboration, Redis and BullMQ for keeping a seperate background worker for PDF and web scraping jobs.<p>I can’t imagine many interior designers are on HN, but figured I would share this for anyone who wants to take a look :) feedback always welcome! The landing page is at <a href="https:&#x2F;&#x2F;inspec.design" rel="nofollow">https:&#x2F;&#x2F;inspec.design</a>

[API/SDK] Show HN: Open-source TypeScript SDK for sending and operating iMessages

[Other] Show HN: FlowLens – MCP server for debugging with Claude Code Hi HN,<p>We often run into this with coding agents like Claude Code: debugging turns into copy-pasting logs, writing long explanations, and sharing screenshots.<p>FlowLens is an MCP server plus a Chrome extension that captures browser context (video, console, network, user actions, storage) and makes it available to MCP-compatible agents like Claude Code.<p>You can try it here: <a href="https:&#x2F;&#x2F;magentic.ai&#x2F;flowlens" rel="nofollow">https:&#x2F;&#x2F;magentic.ai&#x2F;flowlens</a><p>Any feedback—good, bad, or brutal—is welcome.

[Other] /dev/null is an ACID compliant database

[Other] Show HN: OpenSnowcat – A fork of Snowplow to keep open analytics alive I’ve been a long-time Snowplow user and unofficial evangelizer. I have deep respect for its founders, Alex and Yali, who I met a few times.<p>What made me fall in love with Snowplow was that it was unopinionated, gave access to raw event data, and was truly open source. Back in 2013, that changed everything for me. I couldn’t look at GA the same way again.<p>Over the years, analytics moved into SQL warehouses driven by cheaper CPU&#x2F;storage, dbt, reproducibility, and transparency. I saw the need for a democratized Snowplow pipeline and launched a hosted version in 2019.<p>In January 2024, Snowplow changed its license (SLULA), effectively ending open-source Snowplow by restricting production use. When that happened, I realized the spirit of open data and open architecture was gone.<p>A week later, I forked it, I wanted to keep the idea alive.<p>OpenSnowcat keeps the original collector and enricher under Apache 2.0 and stays fully compatible with existing Snowplow pipelines. We maintain it with regular patches, performance optimizations, and integrations with modern tools like Warpstream Bento for event processing&#x2F;routing.<p>The goal is simple: keep open analytics open.<p>Would love to hear how others in the community think we can preserve openness in data infrastructure as “open source” becomes increasingly commercialized.<p>That&#x27;s it, I should have posted here earlier but now felt right.

[Other] Show HN: I built a tech news aggregator that works the way my brain does An honest to god, non-algorithmic reverse chrono list of tech news that passes my signal-to-noise tests, updated hourly.<p>A lightweight a page design as I&#x27;ve been able to keep; simple, clean, fast. No commercial features or aspirations - this is a passion project, something I&#x27;ve been fooling around with on and off for decades.<p>There&#x27;s a &quot;Top&quot; view too with an LLM edited front page &amp; summary, and categorized views for a large number of topics - see the Directory. A few more buried features to explore, but the fundamental use case is pop in, scan, exit - fast and concise.<p>Your feedback would be appreciated!

[Database] Show HN: Pg_textsearch – BM25 Ranking for Postgres I built pg_textsearch, a Postgres extension that brings proper BM25 ranking to full-text search. It&#x27;s designed for AI&#x2F;RAG workloads where search quality directly impacts LLM output.<p>Postgres native ts_rank lacks corpus-aware signals (no IDF, no TF saturation, no length normalization). This causes mediocre documents to rank above excellent matches, which matters when your LLM depends on retrieval quality.<p>Quick example:<p><pre><code> CREATE EXTENSION pg_textsearch; CREATE INDEX articles_idx ON articles USING bm25(content); SELECT title, content &lt;@&gt; to_bm25query(&#x27;database performance&#x27;, &#x27;articles_idx&#x27;) AS score FROM articles ORDER BY score LIMIT 10; </code></pre> Works seamlessly with pgvector or pgvectorscale for hybrid search. Fully transactional (no sync jobs). Preview release uses in-memory architecture (64MB default per index); disk-based segments coming soon.<p>I love ParadeDB&#x27;s pg_search but wanted something available on our managed Postgres. You can try pg_textsearch free on Tiger Cloud: <a href="https:&#x2F;&#x2F;console.cloud.timescale.com" rel="nofollow">https:&#x2F;&#x2F;console.cloud.timescale.com</a><p>Blog: <a href="https:&#x2F;&#x2F;www.tigerdata.com&#x2F;blog&#x2F;introducing-pg_textsearch-true-bm25-ranking-hybrid-retrieval-postgres" rel="nofollow">https:&#x2F;&#x2F;www.tigerdata.com&#x2F;blog&#x2F;introducing-pg_textsearch-tru...</a><p>Docs: <a href="https:&#x2F;&#x2F;docs.tigerdata.com&#x2F;use-timescale&#x2F;latest&#x2F;extensions&#x2F;pg-textsearch&#x2F;" rel="nofollow">https:&#x2F;&#x2F;docs.tigerdata.com&#x2F;use-timescale&#x2F;latest&#x2F;extensions&#x2F;p...</a><p>Feedback welcome, especially from folks building RAG systems or hybrid search applications.

[Other] Antislop: A framework for eliminating repetitive patterns in language models

[Other] Show HN: Git for LLMs – a context management interface Hi HN, we’re Jamie and Matti, co-founders of Twigg.<p>During our master’s we continually found the same pain points cropping up when using LLMs. The linear nature of typical LLMs interfaces - like ChatGPT and Claude - made it really easy to get lost without any easy way to visualise or navigate your project.<p>Worst of all, none of them are well suited for long term projects. We found ourselves spending days using the same chat, only for it to eventually break. Transferring context from one chat to another is also cumbersome. We decided to build something more intuitive to the ways humans think.<p>We started with two simple ideas. Enabling chat branching for exploring tangents, and an interactive tree diagram to allow for easy visualisation and navigation of your project.<p>Twigg has developed into an interface for context management - like “Git for LLMs”. We believe the input to a model - or the context - is fundamental to its performance. To extract the maximum potential of an LLM, we believe the users need complete control over exactly what context is provided to the model, which you can do using simple features like cut, copy and delete to manipulate your tree.<p>Through Twigg, you can access a variety of LLMs from all the major providers, like ChatGPT, Gemini, Claude, and Grok. Aside from a standard tiered subscription model (free, plus, pro), we also offer a Bring Your Own Key (BYOK) service, where you can plug and play with your own API keys.<p>Our target audience are technical users who use LLMs for large projects on a regular basis. If this sounds like you, please try out Twigg, you can sign up for free at <a href="https:&#x2F;&#x2F;twigg.ai&#x2F;" rel="nofollow">https:&#x2F;&#x2F;twigg.ai&#x2F;</a>. We would love to get your feedback!

[Other] Show HN: Nostr Web – decentralized website hosting on Nostr We built Nostr Web, a new way to publish and host websites that live entirely on the Nostr network instead of centralized servers.<p>Each website is a collection of signed, verifiable Nostr events distributed across relays—so it can’t be taken down, censored, or lost.<p>It includes: • DNS TXT records for domain-based discovery (_nweb.domain.com) • CLI publisher tool for versioned deployments (nw-publisher) • Browser extension (nw-extension) for native browsing experience • Relay v1.3.5 support for Nostr Web event kinds<p>Try the live demo: <a href="https:&#x2F;&#x2F;nweb.shugur.com" rel="nofollow">https:&#x2F;&#x2F;nweb.shugur.com</a><p>Repos: <a href="https:&#x2F;&#x2F;github.com&#x2F;Shugur-Network&#x2F;relay" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Shugur-Network&#x2F;relay</a> | <a href="https:&#x2F;&#x2F;github.com&#x2F;Shugur-Network&#x2F;nw-nips" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Shugur-Network&#x2F;nw-nips</a> | <a href="https:&#x2F;&#x2F;github.com&#x2F;Shugur-Network&#x2F;nw-publisher" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Shugur-Network&#x2F;nw-publisher</a> | <a href="https:&#x2F;&#x2F;github.com&#x2F;Shugur-Network&#x2F;nw-extention" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Shugur-Network&#x2F;nw-extention</a><p>Would love feedback from the HN community—on protocol design, relay performance, or UX ideas for improving decentralized web publishing.

[Other] Show HN: Deta Surf – An open source and local-first AI notebook Hi HN!<p>We got frustrated with the fragmented experience of exploring &amp; creating across our file manager, the web and document apps. Lots of manual searching, opening windows &amp; tabs, scrolling, and ultimately copying &amp; pasting into a document editor.<p>Surf is a desktop app meant for simultaneous research and thinking to minimize the grunt work. It’s made of two parts:<p>1) A multi-media library where you can save and organize files and webpages into collections called Notebooks.<p>2) A LLM-powered smart document which you can auto-generate using the context from any stored page, tab or entire notebook. This document contains deep links back to the source material — like a page of a PDF or timestamp in a YouTube video. Unlike Deep Research products (or NotebookLMs chat) the entire thing is editable. The user also stays in the loop.<p>With a technology like AI, context &#x2F; data is proving to be king. We think it should stay under the user’s control, with minimal lock in: where you can own &amp; export, and plug &amp; play with different models. That’s why Surf is:<p>- Open Source on GitHub - Open (&amp; Local Data): the data saved in Surf is stored on your local machine in open and accessible formats and mostly works offline. - Open Model Choice: you can choose which models you use with Surf, and can add custom &amp; Local LLMs<p>Early users include students &amp; researchers who are learning and doing thematic research using Surf.<p>Github repo: <a href="https:&#x2F;&#x2F;github.com&#x2F;deta&#x2F;surf&#x2F;" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;deta&#x2F;surf&#x2F;</a><p>Website: <a href="https:&#x2F;&#x2F;deta.surf&#x2F;" rel="nofollow">https:&#x2F;&#x2F;deta.surf&#x2F;</a>

[Other] Show HN: Subway Surfers in Your Terminal Tired of losing focus when waiting for codegen to finish? Now you can enhance your concentration by integrating subway surfers into your workflow.<p>brew tap evanreilly&#x2F;subway-surfers brew install subway-surfers Usage subway-surfers

[Database] Show HN: SerenDB – A Neon PostgreSQL fork optimized for AI agent workloads Show HN: SerenDB – Neon PostgreSQL fork optimized for AI agent workloads<p>We forked Neon to make database operations faster and safer for AI agents. The goal is to enable instant experimentation with production data and catch prompt injection attacks before they hit your DB. The open-source repo is at <a href="https:&#x2F;&#x2F;github.com&#x2F;serenorg&#x2F;serendb" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;serenorg&#x2F;serendb</a><p>The coolest current features are:<p>1. Time-travel queries: Query your database as it existed at any timestamp. SELECT * FROM orders AS OF TIMESTAMP &#x27;2024-01-15 14:30:00&#x27;. Essential for debugging agent decisions and auditing what data an agent saw.<p>2. Scale-To-Zero with pgvector: Built-in vector embeddings that sleep when idle. Pay nothing for dormant agent databases, auto-scale to 16 vCPU when needed.<p>The coolest features in development are:<p>1. Prompt injection detection: Fingerprinting and context-aware policies to catch injection attempts before they reach your data. We&#x27;re extending Postgres&#x27;s security model to understand AI-specific attack patterns.<p>2. 100ms branch creation: Clone your entire production database in 100ms (vs Neon&#x27;s 500ms). Enables branch-per-agent patterns and instant rollback. Great for faster testing of multiple prompt variations on real data.<p>This is early - we&#x27;re working on the hosted service now.<p>Technical folks: curious if anyone else is hitting the &quot;how do I safely test my agent on prod data&quot; problem. Happy to answer questions about our updates to the branching architecture or our proposed context security approaches.<p>Website: <a href="https:&#x2F;&#x2F;serendb.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;serendb.com&#x2F;</a>

[Other] I built MindMaps for Linux and Git cmdz with quiz, typing game and cheatsheet

[Other] Show HN: Create interactive diagrams with pop-up content This is a recent addition to Vexlio which I think the HN crowd may find interesting or useful.<p>TL;DR: easy creation of interactive diagrams, meaning diagrams that have mouse click&#x2F;hover hooks that you can use to display pop-up content. The end result can be shared with a no-sign-in-required web link.<p>My thought is that this is useful for system docs, onboarding or user guides, presentations, etc. Anything where there is a high-level view that should remain uncluttered + important metadata or details that still need to be available somewhere.<p>You can try it out without signing up for anything, just launch the app here (<a href="https:&#x2F;&#x2F;app.vexlio.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;app.vexlio.com&#x2F;</a>), create a shape, select it with the main pointer tool and then click &quot;Add popup&quot; on the context toolbar.<p>I&#x27;d be grateful for any and all feedback!

[Database] Show HN: Timeplus Proton 3.0 – First vectorized streaming SQL engine Two years after open sourcing Proton, we&#x27;re releasing v3.0 which brings enterprise-grade streaming capabilities to the open source version: full connectivity, processing, and routing in a single binary with zero dependencies.<p>Key features:<p>First vectorized streaming SQL engine in modern C++ with JIT compilation<p>High-throughput, low-latency, high-cardinality processing End-to-end streaming: ETL, joins, aggregation, alerts, and tasks<p>Native connectors: Kafka, Redpanda, Pulsar, ClickHouse, Splunk, Elastic, MongoDB, S3, Iceberg<p>Native Python UDF&#x2F;UDAF support to support your AI&#x2F;ML work loads<p>The same performance we&#x27;ve proven in large enterprise deployments is now available in the community edition.<p>Would love feedback from anyone working with streaming data or looking for Flink&#x2F;ksqlDB alternatives.

[Build/Deploy] SourceFS: A 2h+ Android build becomes a 15m task with a virtual filesystem

[Other] Show HN: Run any GitHub Action locally from your Cron job -- finally!

[Other] Show HN: LunaRoute – a high-performance local proxy for AI coding assistants LunaRoute is a high-performance local proxy for AI coding assistants like Claude Code, OpenAI Codex CLI, and OpenCode. Get complete visibility into every LLM interaction with zero-overhead passthrough, comprehensive session recording, and powerful debugging capabilities.<p>- See Everything Your AI Does - get full logs (JSONL), summary of sessions including tokens used (input&#x2F;output) as well as tools usage and success rates.<p>- Privacy &amp; Compliance Built-In - redact or tokenize any sensitive information (regex based).<p>- Speaks OpenAI and Anthropic dialects so you can route (and translate) when needed between models and providers.<p>- High performance - passthrough is 0.1ms - 0.2ms latency. Logging and summary is off loaded to a secondary thread to allow maximum performance.<p>Feedback is always appreciated as well as stars on the repo :)