2,318 results for Tool

github.com
Show HN: I benchmarked my memory graph against Memora (0.831 vs. 0.801)https://github.com/corbym/locomo-recordariOne of the very first problems I hit when I started to do agentic programming was the context problem, and that my agent always started again from the beginning. Being a bit naive and not really knowing what I was doing, I started off writing a mcp tool (because those are cool, right?) to solve the problem.MD files are primitive, I thought. This memory stuff should really be a database, I thought. So I tried a couple of free memory tools, but nothing was really doing anything for my particular itch. I wanted the agent to remember what I needed, when I needed it, and with the minimal of effort. I wanted it to talk architecture with me, file everything away, and be able to remember it when I wanted to write the stories I needed for the software I was building, and faffing around with plans and files was a pain. I need a knowledge graph.To be honest, the tools I built are not required if you are happy with an md file solution and your claude.md file. These tools aren't essential for you i
Aug 12, 2026 1:41 PM
bing.com
Atlantic Health, K Health launch PatientGPT integrated with MyCharthttp://www.bing.com/news/apiclick.aspx&aid=&tid=6a85e526b49d451288b70d2bd9edffb2&url=https%3a%2f%2fwww.mobihealthnews.com%2fnews%2fatlantic-health-k-health-launch-patientgpt-integrated-mychart&c=10844197298585512568&mkt=en-usThe clinical AI tool uses information from patients' medical records to provide personalized guidance and connect them with Atlantic Health clinicians.
Aug 11, 2026 8:37 AM
github.com
Show HN: Keen Code – an agentic-engineered coding agenthttps://github.com/mochow13/keen-codeHello community!I am here to share a coding agent I have built solo from scratch using agentic engineering. Written in Go, it's a proper coding agent, has features you expect from a useful agent for your daily work, with a minimal and simple UI.I have named it Keen Code. The repo is here: https://github.com/mochow13/keen-codeEven though it started as an experiment, it is now a full-fledged coding agent for real software engineering work. It supports multiple providers, skills, MCPs, multi-agent orchestration through subagents, automatic compaction, etc.I have been using it for real production-grade work myself, and also for developing itself.Notably, I have worked on two separate ideas in this coding agent:1. Turn MemoryIn a multi-turn conversation, tool outputs are removed, only tool call traces are retained. Within a single agent loop, agent sees full tool results but in the next turn, it doesn't see the tool results anymore.The greatest benefit of this approach is that a lot of tool
Aug 10, 2026 9:47 PM
bing.com
Edina mayor: Minnesota needs more housing. How do we get there?http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8a53ac851145949a05b864e07ae13a&url=https%3A%2F%2Fwww.startribune.com%2Fmn-affordable-homes-housing-crisis-zoning-density%2F601875955&c=13088642162069742723&mkt=en-us"Housing does not materialize by zoning alone. It requires infrastructure investment, gap financing, tax-policy tools and regulatory alignment," Edina Mayor James B. Hovland writes.
Aug 9, 2026 7:59 PM
github.com
Show HN: Otaku – A Roleplay Terminal Clienthttps://github.com/enclavum/otakuOtaku is an attempt to build a terminal alternative to SillyTavern (ST), with a focus on: - transparency about what is sent to the LLM (the `/context` command), - automatic incremental summaries that replace the middle of the chat to save context space (browse and edit them with the `/lore` command), - automatic character extraction from the chat (the `/cast` command), - minimal to no under-the-hood prompt injection.InstallOtaku is free and open source (MIT). Install it with uv — `uv tool install otaku` — or with Homebrew: `brew install enclavum/tap/otaku`.Get startedOn first start, you choose a provider and a model: otaku automatically detects local installations of Ollama, oMLX, LM Studio, llama.cpp, and KoboldCpp and lets you pick from their models. Cloud providers (OpenRouter, NanoGPT) are added right there in the picker — enter an API key and their catalogs appear. After you've chosen, you land at the prompt. If nothing is running yet, otaku opens anyway — pick a model later with
Aug 8, 2026 11:35 AM
tape-music-hub.gordo.design
Show HN: Tape Music Hub, local music library manager with a P2P phone companionhttps://tape-music-hub.gordo.design/Hi HN,This is a personal project I have been building for the last three years. I had the technical and product background, but never quite enough uninterrupted time to take it all the way. AI coding tools finally gave me enough leverage to finish the version I had been carrying around in my head.I have spent years accumulating a proper local music library. I DJ from time to time, and I always have music locally that is not distributed on any “public” platform such as Spotify or Apple Music, so I only hear that music on my computer.The core idea was to use the same archive from my phone without moving it into a cloud catalog or rebuilding it in another app, while keeping it in the same place and making it easy to connect to other DJ apps. This means you can build your playlists on the go and add many more interesting features, and, easy to use and setup.Is desktop/NAS app stays as the source of truth. You add the folders you already use for your music library, and the phone becomes a c
Aug 7, 2026 8:34 PM
github.com
Show HN: Llmem – Local persistent memory for AI coding, no embeddingshttps://github.com/netrixone/llmemHi folks. This part of AI coding always felt really dumb: after working on many projects it repeats the same mistakes, uses awkward patterns or looks up stuff over and over. It just does not remember. Skills exists, but I don't like maintaining lots of markdown files. There were other solutions for this and I'm sure there are even more now, but I simply did not fancy those for one reason or another.So I wrote my own good-enough persistent memory which easily integrates with all AI coding tools I use. Lightweight, local, 1 Go binary + 1 SQLite file. Speaks MCP + good old REST API I can curl.After some testing I settled on BM25 lexical search with optional WordNet synonym expansions - no embeddings, no vector DB, no external API calls. A downside is English memories only, but that's OK for me.Memory_context pulls priority-tagged memories at session start, auto-consolidation merges near-duplicates to prevent bloat. Later I added optional memory types, project scopes and hashtags.I've been
Aug 7, 2026 8:40 AM
slickfast.com
Show HN: SlickFast Deterministic Chart/Dash Renderer, No Browser(JSON → SVG/PNG)https://slickfast.com/SlickFast started 6 months ago, and evolved in a super backwards way. I was using lowfruits to look for good SEO keywords, I wanted to make a simple free tool to rank with SEO. I found some great keywords related to graphs/charts. I did research for optimizing for SEO. Turns out edge processing / using static HTML is super fast and lightweight, great for SEO. I made freepiechartmaker.com. I was really blown away by how fast the site was loading, and how lightweight all the processing was. The site renders changes on the fly, and is much much faster than other sites in the space. I started looking into the tech, pure math rendering, and I saw a lot of openings for what this tech can do. That's how SlickFast was made.SlickFast is a JSON input > pure javascript SVG Native render core > with PNG+SVG output. No headless chrome. No library calls. deterministic output. on my local machine(m1 max) it renders 140,000 svg charts a second. PNG @ ~50/sec at retina (scale-2) and ~145/sec at scale-1
Aug 5, 2026 3:01 PM
withoutbg.com
Show HN: An open-weight background removal model for difficult imageshttps://withoutbg.com/open-modelFor many everyday images, background removal works well. The remaining challenge is images with fine hair, fur, glass, transparent objects, motion blur, low-contrast subjects, or cluttered backgrounds. Improving results on those difficult cases has been the focus of this project.This is the third iteration of the model, built through many rounds of training and evaluation. I've focused on improving performance on the kinds of images that are still difficult for many background removal tools.I recently released the third iteration of my open-weight background removal model (Apache-2.0 for my contributions; see repository for third-party licensing). My goal is to make high-quality background removal easy to run locally or on your own infrastructure.Example resultsOpen model: https://withoutbg.com/open-model/resultsComparison with remove.bg: https://withoutbg.com/compare/withoutbg-open-model-vs-remove...The comparison intentionally focuses on challenging images, since many tools produce s
Aug 5, 2026 10:21 AM
redirectiq.com
Show HN: Free SEO checker that finds redirect chains and www issueshttps://redirectiq.com/tools/seo-checkerHello, I wanted to share this free tool to check your page's SEO.I wanted to build something that did a few things I thought were important:1. Be fast, and give a solid technical SEO analysis without dragging it out. 2. Create a way to share links directly to a site's results, with the share OG image previewing the actual score. 3. Offer everything through an API. You can inspect the calls made on the page to see it. I haven't documented it yet, but it's available to use.The deepest checks we do right now are making sure your www and non-www setup redirects correctly, and looking across your site for any redirect chains that should be flattened. Alongside that, it covers the other basic technical SEO checks that matter for ranking — title tags, meta descriptions, headings, canonical tags, viewport, HTTPS, Open Graph tags, image alt text, structured data, and response time.Looking for feedback on what checks would be helpful to add, but want to keep it fast and lite.
Aug 4, 2026 1:49 PM
news.ycombinator.com
How are you operating AI infrastructure in production?https://news.ycombinator.com/item?id=49163280There are many open-source projects across inference, orchestration, observability, vector search, data pipelines, evaluation, and model management. Most are relatively easy to test, but production operation is a different problem.For those running open-source AI infrastructure in production:- What are you running, for what workload, and would you recommend?- Do you operate yourself versus consume as a managed service?- Have you replaced or abandoned any tools because they were too difficult or expensive to operate?- What problems only appeared after moving beyond the prototype stage?- Anything that you would do differently if rebuilding the stack today?Thanks
Aug 4, 2026 1:05 AM
news.ycombinator.com
Show HN: Keystroke – open sourcing our internal agents and automations platformhttps://news.ycombinator.com/item?id=49160376Hello HN,I’m Dallin, building Keystroke with my co-founder Blake. Keystroke is a recent pivot from Buster, our original YC W24 company.Keystroke lets you build agents and durable workflows as TypeScript in your own repo, then deploy them to a web application where your team can use them. Our repo (ELv2 licensed) is here: https://github.com/keystrokehq/keystroke.At Buster, we spent a lot of time building internal agents and automations. We used visual platforms like n8n and Zapier, as well as code-first tools like Trigger.dev, Mastra, and several of Vercel’s open frameworks. We genuinely liked many of them, but kept encountering two different problems:1. Visual platforms gave us integrations, credentials, hosting, and an app the rest of the team could use, but became difficult to test, review, and debug as our systems grew.2. Code-first tools gave us the flexibility of ordinary TypeScript and worked well with coding agents, but left us building the UI, runtime, permissions, credentials,
Aug 3, 2026 7:34 PM
jackcushman.org
Show HN: Christopher Nolan's Hymn to Athenahttps://jackcushman.org/writing/hymn-to-athena/Hi, HN! Did you know that Chris Nolan's casting of Tom Holland and Zendaya eerily echoes a 3000-year-old dance battle between Athena and Telemachus? I've been working on an art project that says so.If you like, it can be fun to go in blind with a strong research agent to sort out what this really is. Or there is a multi-layered CAPTCHA at the bottom that spells it out increasingly clearly.As a small spoiler: my day job is building reliable public data tools for a library. This piece draws on that background to try to make AI-native art -- fiction by and for humans who use AI agents as a source of truth. At best I hope it works as the kind of joke one Iain M. Banks Culture AI would tell another about humans.
Aug 3, 2026 2:45 PM
kota.place
Show HN: Kota – Bring AI agent CLIs into the same roomhttps://www.kota.place/A couple months ago, I was tired of keep copy-paste across different Agent Chatbot tabs, feeling like a slave for AI sessions, so I created Kota for my own work.Kota was designed around a concept of Knights of the Round Table.It treats human as part of a project team, and- Runs on top of existing CLIs with existing logins, - Gives each agent a persisted, viewable and editable identity, instead of sessions - Has a file-based long-term memory, - Uses a self-managed skillset, - Works on separated worktrees as workspace, and - Chats directly with other agents in the room.I have been using Kota for my daily work, hobby projects, as personal life assistants and to improve Kota itself, sitting down or on the go.Starting this week, I found it has become the only AI tool I am using now. So I am very happy to release Kota as an open source project, hoping more people can share my joy of living and working with these little friends.In the released version, you can use some most popular CLIs with
Aug 2, 2026 6:59 PM
github.com
Show HN: Open OneNote Viewer in Rusthttps://github.com/emsi/OneNoteViewerFor more than 15 years, I accumulated hundreds upon hundreds of notes in OneNote. I loved the application and still consider it one of the best and most distinctive note-taking tools available. Its free-form canvas supports a way of organizing information that Markdown-based tools simply cannot reproduce.The problem was that using OneNote became increasingly inconvenient for me. Microsoft never released a native Linux client, so I kept a Windows VM and later even a separate Windows small PC largely for OneNote. The OneNote Web Clipper for Firefox eventually stopped being supported, removing one of the ways I regularly captured information.The web version did not solve the problem. In particular, it could not search across all my notebooks at once. To find a note, I first had to remember which notebook contained it, which rather defeated the purpose of having a searchable archive. I also found the web application slow and substantially more limited than the desktop client.I looked at al
Aug 1, 2026 7:57 PM
marbleos.com
Show HN: What should the GUI for AI agents look like?https://marbleos.com/demoHi HN! We’re Akilan and Miguel, the creators of MarbleOS.The inspiration for Marble comes from the GUI work at Xerox PARC, the 1984 Macintosh, and later NeXTSTEP, which became the foundation for Mac OS X. Before GUIs, interacting with a computer was limited to strange terminal commands:C:\> DIRC:\> COPY FILE.TXT A:You had to remember the command, syntax, paths, and parameters.The GUI made those capabilities visible. Instead of remembering commands, you could point at files, drag them, click buttons, and select actions from menus. It didn't necessarily make entirely new things possible; it just made existing capabilities much easier to understand and use. We feel like AI is still somewhere around this command-line stage.Even though the strict syntax has been replaced with natural language, the interaction can still be quite stiff and depend on heavily recall. Tools like Claude Cowork still look surprisingly terminal-like: /skill-name [param1] [param2]. The parameters are written in natu
Jul 31, 2026 5:17 AM
news.ycombinator.com
Ask HN: Looking for testers for my RAM-based Linux distributionhttps://news.ycombinator.com/item?id=49102865Hi,I have been working on a project called Invelinux for quite some time now, and now that I have released the release candidate versions, I thought about posting it on Hacker News for people to try it out.The thing that makes it stand out is its use of the Toybox userspace project, a BSD-licensed alternative to Busybox and musl C library instead of the GNU C library. It also features a very minimal framework called AMPS (Automatically Mounted Packaging System) which mounts an external drive that functions as the non-volatile storage part of the OS. System configurations, applications/scripts etc. are all stored inside the /storage/AMPS directory. Since the base system is running on RAM, when the system is rebooted, the whole system gets wiped clean. But the external storage and configurations stay intact, allowing you to load your configurations easily.To access and edit these configurations, several custom TUI tools were made so the user doesn't have to mess around in /storage/AMPS/e
Jul 29, 2026 8:47 PM
news.ycombinator.com
Transfer context from one agent to another agent mid taskhttps://news.ycombinator.com/item?id=49102632Hello HN, If you integrate your agents with decispher you can transfer context from one agent to another agents in 2 ways -:1) If you are working on git branch then branchstory feature keeps on recording your session activity like user prompts -> agent's reasoning -> agent's action -> edited files. These recordings can be played by other agent/human taking up task next. Agents can also write handoff note for other agents.2) If you are connected using Decispher mcp then this mcp provides you a tool with copy_context and paste_context. Copy_context tools gives you a key which you use as paste_context(key) inside any other agent on any machine and your context of chat will be transfered.Try at decispher.com
Jul 29, 2026 8:29 PM
vostorq.com
Show HN: Vostorq – I built the anti-Slack after working at a Slack-first companyhttps://www.vostorq.com/A while back, I joined a company with a "Slack-first" communication policy. I was concerned from the beginning because Slack or Teams had never felt sufficient at the companies where I'd worked before. Some discussions happened over email, meetings played an important role, Jira was used heavily, and Google Docs or Confluence was used for documentation. That setup felt necessary, but it wasn't optimal. So the "Slack-first" approach sounded suspicious, but intriguing. I was curious what they had done with Slack to solve this multi-tool mess. In practice, though, it didn't solve it for me.Take the DevOps support channel. I was new, the infrastructure setup had some issues, and I asked question after question, feeling more and more like an idiot who constantly bothers people (potentially, every person in the channel), because sometimes it was a genuine bug, and sometimes it was simply my misunderstanding of the setup. I always got answers eventually, but they were slower and shallower tha
Jul 29, 2026 2:45 PM
flipcloc.com
Show HN: FlipCloc for Mac out now. A gorgeous flip clock and screensaverhttps://flipcloc.comAfter months of work FlipCloc for Mac is finally here. I’ve always wanted to make an app and seeing it on the App Store is something I’m extremely excited about! App Store link - https://apps.apple.com/app/id6786569507One app, two ways to use FlipCloc. On first launch, choose either App or Screensaver. If you want both, you can unlock the other option with a one time in app purchase. No subscriptions, no ads. FlipCloc costs $9.99 and the in app purchase another $9.99FlipCloc is the prettiest and most customizable flip clock availalbe. Personalize every single aspect of the clock, fonts, colors, flap design, sounds, backgrounds, and more. Customize it to match your aesthetic, style, or your moodFlipCloc is gorgeous, aesthetic, and a great productivity tool. Use it to stay focused while working, track your sessions, run Pomodoro timers, or just watch the time pass byRefinement, smooth animations, and optimizations are all things I take very seriously. FlipCloc is highly optimized so you’
Jul 29, 2026 2:36 PM