Ideas Lab
Real problems people complain about online, pulled every morning and scored out of 100. Build, validate, or skip. How scoring works
- Ideas
- 4,820
- BUILD
- 702
- Sources
- 8
- Last sync
- 21h ago
Devthropology – Better Insights for GitHub Repos
Devthropology is a passion project built on top of GitHub pull data. The name is a play on developer anthropology. Pull request data can be cut a lot of ways. The functionality has been built out of curiosity as I want to see different insights into codebases that I work on. Some of the data is typical and other parts I haven't seen elsewhere.I think of this as an improved GitHub Insights page, with faster performance, more detail, and a focus on how work moves through a codebase. The main entity is a contributor, which has two sides: authoring PRs and reviewing/giving feedback to others. From there, you can see repository wide stats, user interactions, contribution trends, file health, and collaboration patterns. Some insights are useful for understanding velocity and code health in the AI era.Details for each page:- Homepage: A high level summary of the repository. Showing age, file types, active contributors, new and churned users. I track the author age at merge, so you can see the tenure of people shipping changes over time.- File explorer: One of my favorite parts. I build a graph of files, tracking renames and moves, to build a complete history. Rolling up, every file and folder is assigned an outlook such as active, developed, stale, touched by people who are likely gone. You can easily see contributor timelines, recent changes, and for some files, their rename/move history and related files that often change together (useful for a coding agent).- Trends: The densest page, showcasing the velocity of contributions and trying to understand if AI is helping ship more. Charts are cut by year for comparisons, tracking PR size, output, rounds of review, and approval latency by different percentiles. PRs are further cut into bucketed sizes to help drill in deeper. Helps to show that smaller PRs are likely still faster to ship while very large PRs (product of AI?) are slowing down.- Relationships: A graph of interactions between contributors, weighted
Hacker News3mo agoToolAI
Easy hosting for personal AI agents like Nous
Need simple, user-friendly hosting options for setting up personal AI agents, prioritizing ease of use over complex setups.
X3mo agoToolAI
LastShelf – an emergency map of your family's documents bills& contacts
After my father was diagnosed with Stage 3 kidney cancer, my family was thrown into a tailspin. Getting second opinions, planning surgery, ensuring insurance coverage, coping with the fear. It was a lot to process.In the middle of dealing with all the medical logistics, I realized none of our family could answer if he: - Had a medical directive? - How to trigger his life insurance policy? - Where is his will and who is the executor? - What bank accounts and credit cards existed? - What bills are not on auto-pay? - When these bills due and how are they paid?That wasn’t solved by password managers or budgeting apps. So I built it.LastShelf: automatically discovers, documents and distributes a map of critical life documents, expenses & contacts in the event of an emergency. Register here: https://www.lastshelf.ai/If you’ve lived through a similar crisis, I really want to hear what would have made the process easier.Anyone who shares their feedback with me will get the first year free. Send a note to support [at] lastshelf.ai
Hacker News3mo agoToolAI
I mapped 8.5M research papers into an interactive atlas
When I read papers, I have to jump between multiple tabs to find the dataset, code, videos, peer reviews, and so on. I tried to fix this with this project.It started as a project just for papers on arXiv, but after its initial success on Twitter (got like 1.9k views: the most I have gotten for a post), I have now expanded it to include other openly available papers from PubMed Central, bioRxiv, medRxiv, and eLife. These papers have been linked with their genes, proteins, diseases, drugs, clinical trials, 3D protein structures, code, and cited and similar papers.This project now has four parts:First, a map. I embedded nearly 8.5M papers (with SPECTER2), ran UMAP for 2D representation, and rendered them as a scatterplot. The dots can be clicked to see brief information about the papers, like an LLM TLDR, key findings, peer reviews, linked entities, and more. The clusters are also labeled, though you might have to zoom in.Second, I built a detailed paper page for each paper. They give you the paper's full text, images, videos, peer reviews (from OpenReview), GitHub links, Hugging Face dataset/model links, clinical trials, genes, diseases, 3D protein structures, cited papers, and similar papers. You can also copy the whole page, including the full paper text and image URLs, as markdown for your LLM.Third, I have released an extension so you can read all this information in your sidebar by clicking "open in Tomesphere" that shows up in arXiv, PMC, bioRxiv, Google Scholar, or medRxiv. I have tried to provide as much information as possible in the extension, though for things like viewing all the images or a 3D protein structure, you might still have to go to the paper page using the link provided in the extension.Fourth, all this data is available for your LLM via MCP. The MCP does have a 50-query free limit (this jumps 10x with signup).Note: this project is still in beta, so papers might have some mismatched information. I am rolling out feedback form
Hacker News3mo agoToolAI
Context.dev (YC S26) – API to get structured data from any website
Hi Hacker News, I’m Yahia. I built Context.dev (https://www.context.dev/) to make it really easy to integrate web data into your products and agents.Here’s a demo video: https://www.tella.tv/video/build-faster-with-context-dev-api...Since it’s an API, here are the docs: https://docs.context.dev/quickstart.You can send us a URL and get back clean Markdown, rendered HTML, screenshots, extracted images, etc.. You can also send us a domain and get company or brand context: name, description, logos, colors, fonts, social links, screenshots, style information, and related metadata. For more custom use cases, you can send a URL plus a JSON Schema and ask us to extract structured data from the site into that shape. For example, you might ask for pricing plans, product categories, office locations, support links, integration partners, or anything else that is visible on the public site.The goal is to give developers the output they actually want. Raw HTML is rarely the useful thing; the useful thing is usually Markdown for a model, JSON for an application, a logo for a UI, or a structured company profile for an agent.Before, I worked at Amazon and Sunrun, and co-founded StockAlarm.io & essense.io, both of which were acquired. Also, I built knifegeek.io, which scraped pocket knives from across the internet and listed them easily. The project is outdated now (coming back soon) but back then it hit the frontpage of hacker news and people seemed to like it: https://news.ycombinator.com/item?id=34604281.Just before Context.dev, I built Brand.dev. The idea was that your software product should automatically know about your customer if they sign up with a corporate email. The API pulled brand data such as logos, backdrops, name, description, industry, and more from the public web and surfaced it to your product to integrate as part of their onboarding experience. That’s worth doing because conversion rates on onbo
Hacker News3mo agoToolAI
Free Mermaid Diagram Editor
I've been slowly adding some new free tools to Moxie Docs (partly for SEO, partly to illustrate some of our feature sets before any commitment) for some reason this mermaid editor one blew up on Google rankings so I figured I'd share in case people find it useful! We also have ADR, AGENTS.md, LLMs, and a few other free tools.
Hacker News3mo agoToolAI
AI-generated content looks too artificial
Creators and users need AI output that doesn't obviously look like it was made by AI, to make it more natural and believable in marketing, writing, and content creation.
X3mo agoToolAI
App update failures blocking access to new AI features
Users are frustrated when app updates fail or are delayed, preventing access to new voice models and features despite availability for others.
X3mo agoToolAI
No intelligent way to automatically select the best LLM for each task
Users struggle to manually choose the optimal LLM based on task type, latency, cost, and performance; need an automated router for better results.
X3mo agoToolAI
My 13-year-old built an ant colony tracker
He's 13 years old. He wanted to track his own ant colonies — growth, feeding, humidity, and other metrics. He built the whole app himself with some help from AI tools; I just helped him deploy it to a server. Would love to hear your feedback!
Hacker News4mo agoToolAI
Kastor – Terraform-style specs for AI agents
Hacker News3mo agoToolAI
Microsoft releases Flint, a visualization language for AI agents
Data visualizations are the bridge between user and data.But building AI agents that can generate visualizations reliably can be very tricky:- simple chart specs can be reliable, but generated charts are often of low quality due to reliance on system defaults; - complex chart specs with explicit details can produce good-looking charts, but they are verbose and agents can struggle with reliabilityWe figured out it is a limitation on the language issue (not just AI capability thing) -- current visualization languages are a bit too low-level for AI agents, requiring them to explicitly make visual decisions that are supposed to be handled by a good compiler. Flint is a visualization intermediate language to address this issue, allow AI agents to solve this last-mile human-agent interaction problem. It provides a simple semantic-type based specification, and contains a layout optimization engine that can produce good-looking charts (filled with derived low-level details) from simple high-level specs. The result is also very human understandable and adaptable. Flint powers data formulator for generating visualizations (another open source project from microsoft https://data-formulator.ai/).Flint is available open source, and we built a MCP server that you can directly plug flint in your favorite agent app to play with data.
Hacker News3mo agoToolAI
Rowboat – Open-source, local-first alternative to Claude Desktop
Claude’s desktop app is brilliant, but for our own daily work we kept wanting it to be less like a chat app and more like a full-fledged work app. Rowboat is our attempt at that, including the ability to build your own work surfaces inside Rowboat (more below).Our repo is https://github.com/rowboatlabs/rowboat, and there’s a demo video here: https://www.youtube.com/watch?v=et5yQABJ3xIIn a previous startup, we built a deep-learning product for enterprise support reps, including teams supporting P&G brands. Models took live notes, suggested replies, and recommended actions while support reps were on calls or handling emails. One lesson stuck with us: it's not enough for the AI to be right, the help has to show up where the work is happening.So we added what we came to call “work surfaces”: dedicated areas for email, meetings, notes, browser, and parallel coding, where the assistant can help inside the workflow itself rather than only through chat:- Email client: Rowboat has a simple email client that sorts incoming emails into important vs. everything else, and pre-creates drafts for important emails. As you edit and send emails, it takes notes on your style, so future drafts get closer to your voice.- Meeting notes: We built a Granola-style local meeting notetaker. Notes are stored as plain Markdown files on your machine. After a meeting, Rowboat feeds the notes back into the knowledge graph and updates the relevant people, project, and topic notes.- Browser: We added a built-in browser, isolated from your main one, where you can log in only to the accounts you want the assistant to help with. The assistant uses browser-use skills to navigate websites.- Parallel coding: The code-mode inside Rowboat lets you spin multiple instances of Claude Code or Codex and either work with them directly or let Rowboat use your work context to orchestrate them. We built an ACP (Agent Client Protocol) client in Rowboat for this.- Notes: Row
Hacker News3mo agoToolAI
Fast, native Mac file manager (filters, fuzzy find, 9 MB, no Electron)
My Downloads folder had been left unkept for a really long time and cleaning it up using Finder was quite cumbersome. So I started creating a simple app to help me filter out and delete or move the files in the folder.It started out very basic and the filtering options genuinely helped me clean out the Downloads folder, then as I thought of more features I would like to see in a file manager I started to add them. Some of the features are:- Fuzzy go to folder/file where you only need to write a few letters of a full path, get suggestions and can jump to the correct path instantly - Hover over a file to preview an image or PDF without opening it - Dual-pane view and tabs that remember selected filters - Command palette to find actions fast - Batch rename (with regex support and presets for common operations), image conversion (HEIC/WebP/AVIF to JPG/PNG), zip creation - Bookmarks, Quick Look, single-click open, keyboard controlAnd many other small quality of life features.I used Claude Code for making the app and the tech stack is .NET/C# with AppKit. The app is compiled to Native AOT so the total app size is only around 9 MB. Since file operations are very important to get right (I don’t want to lose any important files) I put a lot of time into hardening file move/copy/delete operations. Copies are written to a temp file and atomically renamed into place. I also made a dedicated audit of move/copy/delete and verified each operation by hand.Another thing that’s important to me is apps that are privacy first so the only network request the app makes is to check if there’s a new version (this is not done at all for the trial version, the trial expiry date is enforced locally by just checking the date and trusting the user) and the only information that is sent is the current app version.The app is Apple Silicon, macOS 12+ and has a 30-day free trial with a one-time launch price of $19.99 (no subscription).
Hacker News3mo agoToolAI
Halo – open-source, tamper-evident runtime evidence for AI agents
Hi HN, I'm Brian, I spent the last few years at Vanta (YC W18), helping startups and enterprises become compliant and I recently started exploring what that might look like in a post-agentic world.The problem Halo solves is: when a company buys an AI agent from a vendor and gives it access to their data, they have no way to check what the agent did with that data. Vendors may have built observability dashboards and audit logs, but those are editable and partisan. SOC 2 and ISO 27001 audit a company's controls, but controls are less predictive when the software is agentic. TLDR: give an agent the same prompt 50 times, and you get 50 slightly different actions/answers - so the only thing worth auditing in a post-agentic world is what happened at runtime.Halo is an open-source project that produces agent runtime evidence. It's a small recorder that records every action an agent takes (eg. tool calls, model calls, data access, etc), and becomes a record in an append-only log. It's hash-chained, so anyone can re-verify.Run the following command to see a fictional example: uvx --from halo-record halo demo --serve Then, delete a line from one of the .jsonl files and reload, and the report will catch that it's been tampered with.To wire up your own agent, run this line of Python: agent = trace(run_my_agent, profile="my-agent", log="audit.jsonl") Then use this to generate a real report and give it to your customers: halo report audit.jsonl -o report.html Disclaimer: this proves integrity, not completeness (as a self-held chain proves nothing was edited but does NOT prove that nothing was omitted). Catching this requires a witness outside the vendor and is what I'm working on next.Halo is Apache-2.0, contains zero runtime dependencies, and is about 4,300 lines of Python with 125 tests (if you prefer TypeScript, here's that repo: https://github.com/bkuan001/halo-record-ts).Give it a
Hacker News3mo agoToolAI
Scan your AI agents for dangerous capabilities
Hacker News3mo agoToolAI
Pulpie – Models for Cleaning the Web
Hey HN, I'm Shreyash, founder of Feyn. We built Pulpie, a family of Pareto optimal models for cleaning the web. Pulpie strips boilerplate (ads, footers, sidebars) from raw HTML and returns just the main content as HTML or Markdown.We match SOTA extraction quality while being 20x cheaper. Cleaning 1 billion webpages costs $7,900 with Pulpie versus $159,000 with Dripper, the current leading extractor.The gains come from architecture. Today's leading extractors are decoders that generate output one token at a time. Each step reads the full model from memory to produce a single token. Conversely, Pulpie models are encoders. They run one forward pass over the full input HTML and label each block as boilerplate or content. As a result, Pulpie is compute-bound while decoders are memory-bound. Cheaper GPUs have relatively more compute than memory bandwidth. This makes Pulpie easy to run optimally.Here's Pulpie and Dripper cleaning the same pages side by side: https://www.youtube.com/watch?v=ibd-tIiQECo. You can try a side-by-side comparison yourself: https://huggingface.co/spaces/feyninc/pulpieOur motivation for Pulpie came from building a deep research harness. Every search API returns noisy content that contains ads, nav elements, and sidebars. In one instance, an ad for "Gemini on Pixel" slipped into our search results, got passed into LLM context, and ended up in the final answer served to the user. Pretty embarrassing moment for us but it helped us realize how bad data kills model intelligence. We built Pulpie to get clean data for cheap.All models are open source on Hugging Face. You can read about our training process and how to use Pulpie here: https://usefeyn.com/blog/pulpie-pareto-optimal-models-for-cl...Happy to answer any questions!
Hacker News3mo agoToolAI
KiCad in the Browser
KiCad, a PCB EDA suite is now working in a browser, you can try it at the link, there's a demo project or you can bring your own. Firefox is best, Chrome is good, Safari is "working".We’re Emergence Engineering, a dev shop from Hungary, mostly working with rich text editors, CRDTs. PCBJam started as my (Viktor, CTO, ex-electrical engineer) hobby project but as time went on I put more and more energy into it, and a product started shaping up in my head, in the last few months we’ve started to focus on this project a bit more, and this is the first MVP~ish result.This project is a ton of fun, ton of learning, ton of improvements over improvements:- I thought there must be ways to emulate the PCB canvas OpenGL code on the web. And yes, there are a lot of ways, all of them very buggy. Turns out it’s faster to just write WebGL code that works with KiCad’s Graphics Abstraction Layer if you add the right intermediate debugging steps. I (with Claude) implemented the features and compared them to native at every step, then the app loaded up the first time and just worked. I spent weeks hunting weird emulation bugs before that.- There was an old wxWidgets web port as a starting point that helped a lot, bringing it up to the level KiCad needed is a long (and still ongoing) task. Thanks ahilss!- Pthreads on the web: with Emscripten it’s possible to port multithreaded apps (used by DRC, software 3D renderer). A lot of Emscripten features (Asyncify, Pthreads, native exceptions) are in a war with each other, but it’s possible.- Asyncify with native exceptions: Asyncify (used to make the WASM code suspend then call into the JS land, emulating blocking C++ calls by rewriting the WASM directly) is not compatible with native exceptions, even on the latest Binaryen version it can’t suspend inside catch arms. If you write a new Binaryen pass then it can, making the bundle 30-40% smaller and the app load in a second instead of 10.- Optimizing bundle size is a fun game. We
Hacker News3mo agoToolAI
zkGolf – Competitive optimization of formally verified circuits
Zero-Knowledge Proofs (ZKPs) let an untrusted proved show that computation was executed correctly without revealing the inputs to the verifier. However to prove anything, the computation first has to be expressed as a circuit: a system of polynomial equations (constraints) over a finite field. Circuits are the assembly language of zk and every constraint costs prover (and sometimes verifier) time, so production circuits are aggressively hand-optimized.Over the last months, we have been experimenting with writing formal specifications instead and letting LLMs produce the circuits: as long as they could prove that their implementation was correct. It started with SHA-256: we hand wrote a specification in Lean for SHA-256 compression, and then we asked LLMs to write the circuit, targeting R1CS arithmetization and large fields.It took a few hours of work for Opus 4.7, and some light steering into the right direction, but in the end the model came up with a reasonable implementation. We then asked the LLM to aggressively optimize the circuits, by driving down a cost metric of the circuit (number of constraints). We immediately got very promising results, just by asking to come up with optimization ideas, implement them and prove that the new circuit still satisfies soundness and completeness. Sometimes, it came up with unsound optimizations, however, since it could not prove them, it backtracked and got itself back on to the right approach.The result was a (non-deterministic) circuit beating the current, human optimized, state of the art for SHA256 compression. This experience lead us to create "zk.golf" which is an open competition to produce optimized, formally verified circuits to lower the bar for the use of ZKPs and make their application more efficient.Come play (https://zk.golf/llms.txt) and learn about formal verification.
Hacker News4mo agoToolAI