Real problems people complain about online, pulled every morning and scored out of 100. Build, validate, or skip. How scoring works
Coding agents don't have long-term memory.But you do have months of full-fidelity agent transcripts stored on your machine.A simple solution that goes a long way: ingest those transcripts and logs into a structured SQLite database, then search them with ranked text match. Everything is fully local and doesn't require anything fancy like a graph database or hosted memory service.This is the idea behind ctx, a Rust CLI that handles the ingestion and searching.We give our agents a skill that tells them to reference past sessions before working in an area. Usually we do this through an "Agent History Research Subagent" whose job is just to prepare a short brief covering any relevant history before the task begins.A real example: sometimes our test suite runs would fail because disk was full on the runner. The correct approach was to run the cleanup runbook, but the root cause of the failure was not clear to the agents, so they would think it was a test regression and go down the wrong rabbit hole debugging. When the agent searched history, it realized this failure had been encountered before and found the right workaround immediately. That got the agent onto the right cleanup path, and later we improved the log output so the same failure would be clearer next time. It's a boring story, but it's real agent productivity.Another nice use case is quickly generating session transcripts for sharing. You can exclude the noisy intermediate messages, so the transcript shows the important parts of the session more cleanly. Try attaching a session transcript to your next PR so your teammate and their agent can review the provenance and prompting behind the change.If you're up for an additional challenge, ask your agent to "exhaustively review all agent history in this repo and find where the SDLC is struggling or isn't agent-native". Using past sessions to recursively improve the agentic SDLC is a loop that we're using a lot t
Hacker News2mo agoToolAI
I'm a bit annoyed by the feeling that we're kind of stuck when it comes to using LLMs for programming.I use Claude Code and Codex, but I haven't been able to enter flow state like I can when I hand write code.This is kind of ironic to me since AI should be a bicycle for the mind, but right now it feels like a bicycle that just brakes abruptly every couple minutes. I stop, wait, review, prompt again.Is there anyone exploring something fundamentally different than the prompt response loop we have today?I actually think the idea of a tab model is directionally better than prompt response.Would love to hear about any startups, personal experiments, etc.
Hacker News2mo agoToolAI
Hi HN, we are Pietro and Luigi, cofounders of Manufact (https://manufact.com), a cloud for MCP apps and servers. We used to be called mcp-use, and still build open source SDKs for MCP under that name: https://github.com/mcp-use/mcp-use. We did a Show HN about that last year: https://news.ycombinator.com/item?id=44747229.Today we want to tell you about our cloud product, Manufact, which is to mcp-use as Vercel is to Next.js. Manufact is an MCP vertical cloud designed for dev teams putting MCP Apps and servers in production.You can ship, iterate on, test and monitor your MCPs, and get them ready for the store submissions. All with the best developer and agent experience in mind.Here is a demo video of the product: https://www.youtube.com/watch?v=R2rbr5OT9LI.We have been working on MCP since April 2025. Our first focus was making it easy to build agents that could use any MCP server, and a lot of people started using our SDKs. Then the harness revolution kicked off: Claude Code, Claude Cowork, ChatGPT, Codex, OpenCode started shipping agent harnesses that made most standalone agent frameworks redundant. That pushed us to the other side of the connection, the servers. If agents were going to consolidate into a few harnesses, then first-class integration with the rest of a company's systems (i.e. MCP) would become the thing that mattered, so we started building up our server SDKs.Then in succession:1. Oct 2025. ChatGPT Apps SDK. OpenAI brings app UIs to ChatGPT, built on top of MCP and the work of mcp-ui. 2. Late 2025. The stores open. ChatGPT starts accepting app submissions, Claude grows its connector directory with selected partners. 3. Jan 2026. MCP Apps becomes official. SEP-1865 merges as the first MCP extension (io.modelcontextprotocol/ui): one UI standard any host can render.Today, all the major clients fully support MCP and are opening marketplaces of reviewed MCPs that can be one click instal
Hacker News2mo agoToolAI
Hey HN, I’m the creator of Mail Memories. Like many of you, I've had my Gmail address for more than 20 years. A few years ago, I got curious and wanted to see what photos were buried deep in my account. I ended up finding lots of "lost" pictures of old friends, family members, and a ridiculous number of vintage memes.I originally built and launched this as a SaaS, but even with code and policies in place that kept users' photos private, I figured everyone would feel more comfortable with a desktop app.So, I threw out the server architecture and completely rewrote it as a 100% local desktop app for Mac and Windows.How it works now: The app connects directly to Google's server from your computer, processes everything entirely on your system, and saves photos straight to your hard drive.You can download your 50 oldest photos for free (no credit card required) just to see what's in there. If you want to download all the pictures in your account, it's a one-time payment of $29. No subscriptions.If you have an old, pre-2010 Gmail account, definitely give it a spin. You'll be surprised at what you find deep in your archive.I'd love to hear your feedback on the layout, scanning performance, or anything else.TL;DR: I turned my SaaS into a local desktop app (Mac/Windows) that recovers decades of forgotten photos from your Gmail. 100% local, no cloud, no subscriptions, no AI.
Hacker News2mo agoToolAI
We're a team of three friends who have been working with forms and Open Source for a decade, and we joined forces together to create something where we can apply all of our experience.We recently released GolemUI, an Open Source library to generate forms dynamically from JSON definitions, with a typed layer to simplify authoring.This library has a lot to offer. These are the main characteristics:1. A JSON engine. The form is governed by a JSON definition that you can store in a DB, version, diff, or generate it with LLMs as a validated JSON.2. We provide also 28 headless components (and growing) that you can style with CSS variables. We offer APIs so you can drop in Material, Shoelace, or your own components.3. A DX typed authoring layer on top to write forms programmatically, that generates JSON. So you don't have to write it.4. The same definition can render the UI components in React, Angular, Vue, Lit, or Vanilla JS.5. We also have a deterministic MCP that has tools for to validate the model's output, generate JSONs or code, and ensure that the definition returned by the LLM is always valid.You can find more information here:Happy to hear any feedback from you and answer any questions!
Hacker News2mo agoToolAI
Hi all, it’s Greg and Max, founders of Parsewise here (https://www.parsewise.ai/api). Parsewise transforms a bucket of unstructured data into schema compliant data, retaining lineage for values resolved across documents.Imagine giving Claude a bunch of files and asking for a CSV or JSON output. If you have tried this, you know both the system limitations (number of files, type of inputs, cost, latency) but also the human-facing challenge of having no way to validate the results quickly. We solve both. We help tech teams simplify their unstructured data ETL, and loop in business experts for the definitions and for instant validation.Here is a video with a few use cases: https://www.youtube.com/watch?v=dbRllnnh47wParsewise in the words of someone coming to us: ”I need to extract information from insurance policy PDFs, phone calls that have been transcribed, emails, etc. I am NOT looking for something that would just extract data point by data point, page by page into a structured well-defined schema but more something more agentic that can understand that information might be across documents and that it should reason over what to extract.”We started the company based on a decade of experience (and pain) in complex data transformation and data analysis / synthesis. Greg was building both classical ETL and implemented AI workflows at Palantir. At Bain, Max did highly complex data analysis in the financial sector, similar to many of our customers.Parsewise works by taking in a bucket of data (think hundreds or thousands of pdfs, excels etc.), and outputting schema compliant data where every single value is traceable down to word level citations across multiple documents in the bucket. We provide API customers with ways to show the lineage in their own applications, or they can use our platform for internal operations. At the core of the data processing we have self-improving agent definitions. They define the acceptable sources, the lo
Hacker News2mo agoToolAI
I believe someone recently posted sort of a secure harness/wrapper for running coding agents in a secure sandbox. I can't find the project.Of course I can make my own wrapper with systemd-nspawn, kata or bspawn, but I believe I saw a decently well-maintained project just a while back. Does anyone have a suggestion or link? It's become extremely hard to find things on GitHub with all the generated projects.
Hacker News2mo agoToolAI
Over the past few months I've been heavily involved in the decompilation community. I've been hands-on decompiling a beloved game from my childhood (Star Fox Adventures). I started this journey with zero prior decomp experience—and to make things worse I had never really touched C nor assembly either.Learning how to decompile was challenging. It's difficult to find any good learning resources for it and any open-source projects for this are inactive and/or contain little actual learning material.So I put together Decomp Academy! Decomp Academy is an interactive way to learn how to decompile PowerPC assembly back into C. The site runs a live Metrowerks CodeWarrior GC/2.0 compiler, converts your C into assembly, and then checks how close your assembly matches the target. If even 1 instruction or bit is off, that's a fail. This is the gold standard for video game decompilation and this is much stricter than a normal decompile.As of writing there are 250+ lessons on the site and the lessons start at the very basics so anyone with a little programming experience should be able to jump straight in, even if you're not a C expert. Some lessons also have real functions taken from live open source decomp projects (Star Fox Adventures, Mario Party 4, Pikmin, Metroid Prime). The idea being you learn everything you need to know to be able to jump in and contribute to a real decompilation project when done.The site is completely free, open source and you have access to all lessons without having to sign up. All lessons are stored in markdown in the repo (src/curriculum), it's trivial to add or modify lessons. The site is very new and the lessons are rapidly changing every day with a whole C++ section on the way. The site has already been well received by the decomp community and I'm happy to share it with HN. I'm very keen on others to contribute to this project and I hope this becomes the best resource on the internet for le
Hacker News2mo agoToolAI
Bash4LLM is a single-file Bash wrapper for interacting with LLMs from the terminal. I created it because I wanted something simple that worked without installing Python, Node, or any other runtime.It uses only Bash, curl, and jq. You can send prompts, start a small chat, process files line by line, stream output, and save session metadata in JSON format.I tried to make it safe and predictable: no use of the system /tmp, no use of eval. Groq is supported by default, and other providers can be added with dedicated Bash scripts in the extras/providers/ folder.Example: echo "explains the command: ls -l" | ./bash4llm
Hacker News2mo agoToolAI
I worked with a few employers on contracts, and there were situations where they haven't paid fees for a while.Is there a list or site that lists all the employers and if they have a record of them not fulfilling their contracts?
Hacker News2mo agoToolAI
A month ago there was a wave of posts and tweets about engineers walking around cafes and parks with their MacBooks propped half-open, as fully closing the lid forces sleep that stops their AI agents. Some people made snarky comments about using tmux or Amphetamine, and some defended their choice with “but I only need it sometimes, and forgetting to disable Amphetamine and finding my laptop discharged in my bag is worse.”This is a solution to this problem. Unlike caffeinate, it will prevent your MacBook from sleeping even with the lid closed, with no external power or display, using pmset disablesleep 1. Unlike other sleep-preventing apps, Adrafinil only activates when there’s an agent actively doing something. It detects agent activity through hooks it installs into Claude Code, Codex, and others. To reassure you it’s working, the app shows the active status in the menu bar, and it plays a chime when you close the lid.Once the agent is done, Adrafinil detects it and lets the laptop go to sleep by setting pmset disablesleep back to 0. It will also let it sleep in case of overheating. And if you want to manually toggle it, you can install an optional MCP and tell your agent to keep the MacBook awake for a specific time.It has four binaries, one of which is a root helper exposing a single setSleepBlocked call. All the logic and policy live in the unprivileged parts. They’re all notarized, and the app is fully open source (MIT).
Hacker News2mo agoToolAI
### What variant of Codex are you using? App ### What feature would you like to see? I love the new codex app. However due to [this issue](https://github.com/openai/codex/issues/10432) it is almost unusable on my mac laptop. I would like to use the app on my linux desktop where power consumption is far less of an issue. ### Additional information _No response_
GitHub7mo agoToolAI
This is a tool to manage encrypted local backups of secrets. The core idea is that it aims to be usable without depending on it, meaning that even if the software disappeared from the face of Earth tomorrow, your data would still be recoverable.It also integrates nicely with NixOS (which is what I use, though it does not require NixOS to be used).I have summed up a bit of explanation and some answers to reasonable questions in a blog post: https://baldino.dev/blog/secs-man/
Hacker News2mo agoToolAI
Made this in a free evening. Index an permissive license translation of the Bible (WEB) into a RAG database to allow returning passages of similar semantic meaning. Lots of fun. For example, "more money more problems" returns Ecclesiastes 5:9-13 which, I'll just say, is spot on.."Moreover the profit of the earth is for all. The king profits from the field. He who loves silver shall not be satisfied with silver, nor he who loves abundance, with increase. This also is vanity. When goods increase, those who eat them are increased; and what advantage is there to its owner, except to feast on them with his eyes? The sleep of a laboring man is sweet, whether he eats little or much; but the abundance of the rich will not allow him to sleep. There is a grievous evil which I have seen under the sun: wealth kept by its owner to his harm."Anyway - thought it was fun enough to share. It's slow and I vibe coded it so I haven't sorted out how to make it not take 15 seconds to vector search against the full 4GB index.
Hacker News2mo agoToolAI
From the official information online, joining the e-Residency program and setting up an Estonian company seems relatively straightforward.I'm considering using an Estonian OÜ for a SaaS business and would love to hear from people who have actually gone through the process.What surprised you after becoming an e-resident and establishing your company? Were there recurring costs, compliance requirements, banking/payment issues, tax complications, or other operational challenges you didn't anticipate?How easy has it been to run the company remotely? Have you ever needed to travel to Estonia to resolve an issue?Looking back, would you do it again? What do you wish you had known before getting started?I'd appreciate any lessons learned, gotchas, or advice from current or former e-residents.
Hacker News2mo agoArticleAI
We built a model router that plugs into coding agents (e.g. Claude Code, Codex, Cursor, etc.) and intelligently sends requests to the best model to serve them. Here's a quick demo of running it locally: https://www.youtube.com/watch?v=isKhAyivtfM.At Weave, we write most of our code with AI, and it's been getting more expensive. This came to a head when Opus 4.7 was released and, thanks to its tokenizer changes, our costs shot up. We knew we didn't need Opus for everything but we didn't want to lose out on the intelligence for the cases where you really need it. So we decided to build a model router to handle this for us.The Weave Router acts as an Anthropic/OpenAI endpoint specifically for coding agents. It looks at every inference request and intelligently (more on that in a sec) decides what model to send it to, handling all the translations required along the way. So it can use faster/cheaper models (e.g. DeepSeek v4, GLM 5.2, Kimi K2.6) when possible, and frontier models (Opus 4.8 & GPT 5.5 (& Fable whenever it's back)) when necessary.How do we know what model to route to? We trained an RL model on tens of thousands (so far!) of agent traces. We reward the routing model when it selects an LLM that successfully completes the given task.Here's an example: if you ask the router to plan a complex change, it will (probably) route that request to Opus 4.8. Subagents exploring the codebase to gather context will be routed to more suitable models (e.g. DeepSeek V4 Flash). Then when you have the plan ready to implement, it will be (most likely) be handed to a quicker model (e.g. GLM 5.2) to carry it out.We've been using this internally for the last month or so. We've saved 40% on tokens vs. what we otherwise would have paid, with no noticeable differences in quality or velocity.The router is source-available under Elastic License 2.0, so you can self-host it. Or if you prefer, you can also use our h
Hacker News2mo agoToolAI
I built an experiment that uses an overfitted transformer and arithmetic coding to compress individual files.Instead of training the model to generalize, I train a 900KB transformer to memorize a single file and predict the next byte. Those predictions are fed into an arithmetic coder to produce the compressed output.On a 100MB NYC taxi CSV, it compresses to about 7MB (~0.5 bits/byte). On a 100MB slice of enwik9, it compresses to about 21MB (~1.68 bits/byte).It's pretty slow right now (roughly 20–30 minutes of training and 45 minutes each for compression and decompression on my AMD 7800XT).Checkout the repo - https://github.com/samyak112/pym-particles
Hacker News2mo agoToolAI
Hacker News3mo agoToolAI
Hi HN, Nick here. We’re launching OpenKnowledge (https://openknowledge.ai/), a “what you see is what you get” markdown editor that has direct integrations with Claude, Codex, and other agents. Available as MacOS app or Web UI+CLI. Fully free/local and OSS.We built this because we wanted a Notion-like experience for writing and sharing markdown files across our team. Obsidian is the best alternative we tried, but found it doesn’t have a true WYSWIG UI and it didn’t integrate well with Claude/Codex outside of community plugins.So we built OpenKnowledge. It takes shape as:1. A MacOS app with a file navigator, the WYSIWYG editor, and link explorer.2. Integrations with the Claude, Codex, and Cursor desktop apps. The agents can open an OpenKnowledge editor within their embedded web browsers for a side-by-side experience.3. Built-in mcps, skills, and RAG for LLM-wiki and “AI Second Brain” scenarios + spec writing4. An embedded terminal and CLI for TUI-first usersOSS stack includes: Tiptap/prosemirror, CodeMirror, yjs (CRDT), Electron (MacOS app), Orama, remark/rehype/micromark/mdast, @pierre/treesOn the architecture side, the interesting eng. challenges included:1. A pipeline to convert ProseMirror to markdown in a bidirectional lossless way. ProseMirror uses ASTs, which are not designed to have byte-fidelity.2. A dual-observer CRDT to keep the ProseMirror and markdown state in-sync.The CRDT + git also power a collaborative experience that shows what Agents are doing in the markdown, have undo/redo, and version history. The “Share” and cloud-sync functionality are geared for team collaboration. They feel “no-code” but leverage git/GitHub under the hood, which also means data stays fully private.In that spirit, we made OpenKnowledge open source for anybody who’s curious or who’d like to contribute.We’re actively thinking about plugins/extensibility and what’s next. If you have suggestions or feedback, would lo
Hacker News2mo agoToolAI
Logged into my old email after a year - it has everything setup with recent details, so I know for a fact it isn't more than a year. To my shock, all my data is now gone. Just a simple `fuck you` email at the top of my inbox. Is there any way to recover my old emails? Unfortunately I used to use this as my main account since the early 2000s. No shady stuff, just your average joe with an average email. Here's the only email now in my account:Dear Yahoo Mail Member,We’re so glad to see you’re back! Due to inactivity, your mailbox entered an inactive state. This means we stopped accepting your incoming emails. If your period of inactivity was long enough, we also deleted all of your emails. Now that you’ve signed back in, your mailbox has been reactivated and you will start receiving new emails again shortly. For more information, please visit the Yahoo Mail policy page at: https://en-global.help.yahoo.com/kb/mail/SLN2018.htmlWe’ve made some upgrades to your email experience since the last time you signed in. Now Yahoo Mail has a cleaner design and upgraded features to help you keep your inbox in order. Find whatever you need quickly without the clutter. Search by keyword, contact, date or any combination of these. With themes, you can pick your favourite colour and the level of intensity. The choice is yours. Stay on top of everything on the go. The top-rated Yahoo Mail app makes life easy to manage with features such as Subscriptions View. See all the mailing lists you’re subscribed to and easily unsubscribe with a single tap. You can download it here: https://mail.onelink.meWe hope that you enjoy your new Yahoo Mail experience.Thank you,The Yahoo Mail team
Hacker News2mo agoToolAI