Real problems people complain about online, pulled every morning and scored out of 100. Build, validate, or skip. How scoring works
I've spent the past few years building 50+ AI agents in prod (some reached 1M+ sessions/day), and the hardest part was never building them — it was figuring out why they fail.AI agents don't crash. They just quietly give wrong answers. You end up scrolling through traces one by one, trying to find a pattern across hundreds of sessions.Kelet automates that investigation. Here's how it works:1. You connect your traces and signals (user feedback, edits, clicks, sentiment, LLM-as-a-judge, etc.) 2. Kelet processes those signals and extracts facts about each session 3. It forms hypotheses about what went wrong in each case 4. It clusters similar hypotheses across sessions and investigates them together 5. It surfaces a root cause with a suggested fix you can review and applyThe key insight: individual session failures look random. But when you cluster the hypotheses, failure patterns emerge.The fastest way to integrate is through the Kelet Skill for coding agents — it scans your codebase, discovers where signals should be collected, and sets everything up for you. There are also Python and TypeScript SDKs if you prefer manual setup.It’s currently free during beta. No credit card required. Docs: https://kelet.ai/docs/I'd love feedback on the approach, especially from anyone running agents in prod. Does automating the manual error analysis sound right?
Hacker News5mo agoToolAI
Libretto (https://libretto.sh) is a Skill+CLI that makes it easy for your coding agent to generate deterministic browser automations and debug existing ones. Key shift is going from “give an agent a prompt at runtime and hope it figures things out” to: “Use coding agents to generate real scripts you can inspect, run, and debug”.Here’s a demo: https://www.youtube.com/watch?v=0cDpIntmHAM. Docs start at https://libretto.sh/docs/get-started/introduction.We spent a year building and maintaining browser automations for EHR and payer portal integrations at our healthcare startup. Building these automations and debugging failed ones was incredibly time-consuming.There’s lots of tools that use runtime AI like Browseruse and Stagehand which we tried, but (1) they’re reliant on custom DOM parsing that's unreliable on older and complicated websites (including all of healthcare). Using a website’s internal network calls is faster and more reliable when possible. (2) They can be expensive since they rely on lots of AI calls and for workflows with complicated logic you can’t always rely on caching actions to make sure it will work. (3) They’re at runtime so it’s not interpretable what the agent is going to do. You kind of hope you prompted it correctly to do the right thing, but legacy workflows are often unintuitive and inconsistent across sites so you can’t trust an agent to just figure it out at runtime. (4) They don’t really help you generate new automations or help you debug automation failures.We wanted a way to reliably generate and maintain browser automations in messy, high-stakes environments, without relying on fragile runtime agents.Libretto is different because instead of runtime agents it uses “development-time AI”: scripts are generated ahead of time as actual code you can read and control, not opaque agent behavior at runtime. Instead of a black box, you own the code and can inspect, modify, version, and debug
Hacker News5mo agoToolAI
Vector databases store memories. They don't manage them. After 10k memories, recall quality degrades because there's no consolidation, no forgetting, no conflict resolution. Your AI agent just gets noisier.YantrikDB is a cognitive memory engine — embed it, run it as a server, or connect via MCP. It thinks about what it stores: consolidation collapses duplicate memories, contradiction detection flags incompatible facts, temporal decay with configurable half-life lets unimportant memories fade like human memory does.Single Rust binary. HTTP + binary wire protocol. 2-voter + 1-witness HA cluster via Docker Compose or Kubernetes. Chaos-tested failover, runtime deadlock detection (parking_lot), per-tenant quotas, Prometheus metrics. Ran a 42-task hardening sprint last week — 1178 core tests, cargo-fuzz targets, CRDT property tests, 5 ops runbooks.Live on a 3-node Proxmox homelab cluster with multiple tenants. Alpha — primary user is me, looking for the second one.
Hacker News5mo agoToolAI
Companies building software rapidly with AI overlook basic UX issues, resulting in frustrating products that customers endure despite dissatisfaction.
X6mo agoToolAI
AI coding assistants like Grok provide supposedly complete code but omit sections, make undocumented changes, and introduce new errors while fixing others.
X5mo agoToolAI
As AI agents evolve into digital workers, they require dedicated authentication systems, session management, and access control to securely log into services like humans do.
X5mo agoToolAI
MSMEs in India rely on basic tools like WhatsApp and Excel because enterprise AI solutions are prohibitively expensive (lakhs per month), needing practical affordable automation.
X5mo agoToolAI
are you ok sharing secrets or api keys to you ai agent via .env?or is there any other tool or mechanism that one use to safegaurd from potential exploit or leaks
Hacker News5mo agoToolAI
Hi folks, today we're launching QVAC SDK [0], a universal JavaScript/TypeScript SDK for building local AI applications across desktop and mobile.The project is fully open source under the Apache 2.0 license. Our goal is to make it easier for developers to build useful local-first AI apps without having to stitch together a lot of different engines, runtimes, and platform-specific integrations. Under the hood, the SDK is built on top of QVAC Fabric [1], our cross-platform inference and fine-tuning engine.QVAC SDK uses Bare [2], a lightweight cross-platform JavaScript runtime that is part of the Pear ecosystem [3]. It can be used as a worker pretty much anywhere, with built-in tooling for Node, Bun and React Native (Hermes).A few things it supports today: - Local inference across desktop, mobile and servers - Support for LLMs, OCR, translation, transcription, text-to-speech, and vision models - Peer-to-peer model distribution over the Holepunch stack [4], in a way that is similar to BitTorrent, where anyone can become a seeder - Plugin-based architecture, so new engines and model types can be added easily - Fully peer-to-peer delegated inference We also put a lot of effort into documentation [5]. The docs are structured to be readable by both humans and AI coding tools, so in practice you can often get pretty far with your favorite coding assistant very quickly.A few things we know still need work: - Bundle sizes are larger than we want right now because the current packaging of Bare add-ons is not as efficient as it should be yet - Plugin workflow can be simpler - Tree-shaking is already possible, but at the moment it still requires a CLI step, and we'd like to make that more automatic and better integrated into the build process This launch is only the beginning. We want to help people build local AI at a much larger scale. Any feedback is truly appreciated! Full vision is available on the official website [6].References:[0] S
Hacker News5mo agoToolAI
https://pardonned.comInspired by the videos of Liz Oyer, I wanted to be able to verify her claims and just look up all the pardons more easily.Tech Stack: Playwright - to sccrape the DOJ website SQLite - local database Astro 6 - Build out a static website from the sqlite dbAll code is open source and available on Github.
Hacker News5mo agoToolAI
Hi HN. I built Waffle because I kept ending up with 15 terminal windows scattered across three spaces with no idea what was running where.Splitting/merging in iTerm kind of works but it never felt intuitive to me.With that in mind, I built something to suit my workflow:Waffle is a native macOS terminal (Built on Miguel de Icaza's SwiftTerm) that tiles your sessions into an auto-scaling grid automatically. 1 session is fullscreen, 2 is side by side, 4 is 2x2, 9 is 3x3. Open a terminal, it joins the grid. Close one, the grid rebalances. No splitting, no config.I've been using it a lot recently and one thing I've found really useful is that sessions detect which repo they're in and group accordingly. Each project gets a distinct colour. Cmd+[ and Cmd+] flip between groups. If you have three repos open across eight terminals, you can filter to just one project's sessions instantly. Also, no accidentally closing a window with CMD-W as it gives you a confirmation and requires a second CMD-W to close.Honestly, if you live in tmux, this probably isn't for you but it's really helped to speed up my workflow.Other things: It comes with a handful of themes (and has support for iTerm themes), bundled JetBrains mono, has keyboard shortcuts for everything. Free, no account, opt-in analytics only. macOS 14+.There's a demo on the landing page if you want to see it in action.
Hacker News5mo agoToolAI
Hello everyone.Claudraband wraps a Claude Code TUI in a controlled terminal to enable extended workflows. It uses tmux for visible controlled sessions or xterm.js for headless sessions (a little slower), but everything is mediated by an actual Claude Code TUI.One example of a workflow I use now is having my current Claude Code interrogate older sessions for certain decisions it made: https://github.com/halfwhey/claudraband?tab=readme-ov-file#s...This project provides:- Resumable non-interactive workflows. Essentially `claude -p` with session support: `cband continue <session-id> 'what was the result of the research?'` - HTTP server to remotely control a Claude Code session: `cband serve --port 8123` - ACP server to use with alternative frontends such as Zed or Toad (https://github.com/batrachianai/toad): `cband acp --model haiku`. - TypeScript library so you can integrate these workflows into your own application.This exists cause I was using `tmux send-keys` heavily in a lot of my Claude Code workflows, but I wanted to streamline it.
Hacker News5mo agoToolAI
Hi HN!I recently switched from a Fedora/GNOME laptop to a MacBook Air. My old setup served me well as a portable workstation, but I’ve started traveling more while working remotely and needed something with similar performance but better battery life. The main thing I missed was a simple taskbar that shows the windows in the current workspace instead of a Dock that mixes everything together.I built boringBar so I would not have to use the Dock. It shows only the windows in the current Space, lets you switch Spaces by scrolling on the bar, and adds a desktop switcher so you can jump directly to any Space. You can also hide the system Dock, pin apps, preview windows with thumbnails, and launch apps from a searchable menu (I keep Spotlight disabled because for some reason it uses a lot of system resources on my machine).I’ve been dogfooding it for a few months now, and it finally felt polished enough to share.It’s for people who like macOS but want window management to feel a bit more like GNOME, Windows, or a traditional taskbar. It’s also for people like me who wanted an easier transition to macOS, especially now that Windows feels increasingly user-hostile.I’d love feedback on the UX, bugs, and whether this solves the same Dock/Spaces pain for anyone else.P.S. It might also appeal to people who feel nostalgic for the GNOME 2 desktop of yore. I started my Linux journey with it, and boringBar brings back some of that feeling for me.
Hacker News5mo agoToolAI
Agents forget 'what happened, when, why it mattered,' limiting complex tasks; builders publicly developing persistent memory solutions.
X5mo agoToolAI
AI agents built by developers get blocked by websites because they originate from data center IPs, preventing realistic browsing.
X5mo agoToolAI
AI agents cannot retain knowledge or learn from previous sessions, forcing developers to restart context every time.
X5mo agoToolAI
I built this because I was interested in the data. Didn't fully get it to what I wanted, but thought I'd share it nonetheless. Maybe someone has better data sources they could share!Turns out live ship tracking APIs are expensive so I manually just copied the json from https://www.marinetraffic.com/en/ais/home/centerx:57.4/cente... I'll probably have an ai agent do the same thing on some cron interval, if this gets any fanfare.To actually know if the port is open without live ship tracking I found https://portwatch.imf.org/pages/cb5856222a5b4105adc6ee7e880a... which was perfect, except it has 4 day lag!I also thought of adding news feed parsing or prediction market data to get a more definitive answer on if it's open right when you load it, but I spent a few hours and am gonna move on for now.
Hacker News5mo agoToolAI
Developers are seeking robust tools beyond LangGraph to manage and orchestrate multiple AI agents independently without constant oversight.
X5mo agoToolAI
Eve is an AI agent harness that runs in an isolated Linux sandbox (2 vCPUs, 4GB RAM, 10GB disk) with a real filesystem, headless Chromium, code execution, and connectors to 1000+ services.You give it a task and it works in the background until it's done.I built this because I wanted OpenClaw without the self-hosting, pointed at actual day-to-day work. I’m thinking less personal assistant and more helpful colleague.Here’s a short demo video: https://www.loom.com/share/00d11bdbe804478e8817710f5f53ac61The main interface is a web app where you can watch work happen in real time (agents spawning, files being written, use of the CLI). There's also an iMessage integration so you can fire a task asynchronously, put your phone down, and get a reply when it's finished.Under the hood, there's an orchestrator (Claude Opus 4.6) that routes to the right domain-specific model for each subtask: browsing, coding, research, and media generation.For complex tasks it spins up parallel sub-agents that coordinate through the shared filesystem. They have persistent memory across sessions so context compounds over time.I’ve packaged it with a bunch of pre-installed skills so it can execute in a variety of job roles (sales, marketing, finance) at runtime.Here are a few things Eve has helped me with in the last couple days:- Edit this demo video with a voice over of Garry: https://www.youtube.com/watch?v=S4oD7H3cAQ0- Do my tax returns- To build HN as if it was the year 2030: https://api.eve.new/api/sites/hackernews-2030/#/AMA on the architecture and lmk your thoughts :)P.S. I've given every new user $100 worth of credits to try it.
Hacker News5mo agoToolAI
Hello HN users,This is a CAD by code project I have been working on on my free time for more than year now.I built it with 3 goals in mind:- It should be familiar to CAD designers who have used other programs. Same workflow, same terminology.- Reduce the mental effort required to create models as much as possible. This is achieved by: - Provide live rendering and visual guidance as you type. - Allow the user to reference existing edges/faces on the scene instead of having to calculate everything. - Provide interactive mouse helpers for features that are hard to write by code: Only 3 interactive modes for now: Edge trimming, Sketch region extrude, Bezier curve drawing. - Implicit coding whenever possible: e.g: There are sensible defaults for most parameters. The program will automatically fuse intersecting objects together so you do not have to worry about what object needs to be fused with what. - It should be reasonably fast: The scene objects are cached and only the updated objects are re-computed.I think I have achieved these goals to a good extent. The program is still in early stages and there are many features I want to add, rewrite but I think it is already usable for simple models.Update to add more details: This is based on Opencascade.js WASM binding. So you get all the good things that come with any brep kernel. Fillets, chamfers, step import and export...The scene is webview but the editing is in your local file. You use your own editor and the environment you are familiar with.One important feature that I think make this stand out among other code based cad software is the ability to transform features not just shapes. More here: https://fluidcad.io/docs/guides/patterns You can see it in action in the lantern example: https://fluidcad.io/docs/tutorials/lantern
Hacker News5mo agoToolAI