Indexes - Moozonian Search

github.com• Jul 28, 2026• 1 min read

Show HN: FedTerm – a Claude-native terminal for macOS that popup like Spotlighthi hn, i made a terminal for mac. you press a hotkey and it pops up over whatever window you're in, kinda like spotlight. press it again and it's gone.it saves and indexes all your claude code sessions, so you can search through them and pick up any one right where you left it. the same window also searches your shell history and your saved ssh hosts.swift + swiftterm, mit licensed, no accounts, nothing to sell. happy to hear what you think.github.com/feddot2517/fedterm

news.ycombinator.com• Jul 24, 2026• 1 min read

Ask HN: Multi dimensional sort (beyond Hilbert curve)My point is about structuring and indexing an arbitrary dataset, for preprocessing before eg segmentation/clustering, neighbor querys...In 1D, sorting is trivial because scalar values have a total, unambiguous order. In 2D and higher dimensions, there is no single way to order points, and points can't all be compared >=/=_k v if u_k >= v_k (comparing along the k-th coordinate axis only).In 1D, a sorted array satisfies x[i+1] >= x[i]. Extending this naturally to a multi-indexed d-dimensional array x[i, j, k, ...], we can define a dataset as "cartesian sorted" if:- x[i+1, j, k, ...] >=_1 x[i, j, k, ...] - x[i, j+1, k, ...] >=_2 x[i, j, k, ...] - ...and so on.In other words, the array of multi indexes is sorted along every row, column, and slice under its corresponding coordinate projection.What makes this definition interesting is that their exist a very simple algorithm to perform Nd sort in the "cartesian" sense. Just iterate: 1. Perform 1D sorts along every row (apply the permutation

news.ycombinator.com• Jul 19, 2026• 1 min read

Prodigy – AI Workforce for professional teamsHello folks! I am Samay, founder of Prodigy. We're building an AI infrastructure that provides on-demand subagents who work like autonomous teammates alongside your team.How are we any different? 1. Our Prodigy brain indexes every bit of information from tools used by your company. This includes emails, documents, meetings, conversations, git work, CRM etc. The agents communicate with the brain instead of humans.2. The Prodigy orchestrator spins up agents with designated roles and goals. Each agent has a clear objective, evidence, verification tests and stopping condition.If you're interested about the product or want to chat in general, please reach out! Website: https://www.prodigy.org.in

bing.com• Jul 17, 2026• 1 min read

Stock market today: Dow, S&P 500, Nasdaq post weekly losses as semiconductors get smokedUS stocks declined on Friday, with the major indexes posting weekly losses led by declines in the semiconductor sector. The S&P 500 dropped more than 1.5% for the week while the Nasdaq declined 2.9%.

contextify.sh• Jul 3, 2026• 1 min read

Show HN: Pull Claude Code transcripts into your Codex session, and vice versaHi HN, I'm Rob. Contextify indexes every Claude Code and Codex session on your machine into one local, searchable database. The current session, in either tool, can search all of it: /total-recall in Claude Code, $total-recall in Codex.Demo: https://www.youtube.com/watch?v=FvrvRGp4C9M | Mac app: https://contextify.sh (App Store or DMG) | Linux: CLI with a one-line installer, same search. No signup.I split my work between Claude Code and Codex. When I burn through rate limits on one, I switch to the other, and new models keep leapfrogging each other, so the switching is not going away.But each tool keeps its own history, in its own format, in its own directory structure, and Claude Code deletes local transcripts after 30 days by default.Piecing together work I had done across Claude Code and Codex on one machine is the problem Contextify started as.How it works: the Mac app (or a systemd timer on Linux) watches ~/.claude/projects/ and ~/.codex/sessions/, parses each tool's transcript fo

news.ycombinator.com• Jun 20, 2026• 1 min read

Forked CozoDB to give agents cognitive primitivesCozoDB is an embedded relational-graph-vector database created by Ziyang Hu and the Cozo Project authors. It does Datalog, vector search (HNSW), and full-text search in a single embedded engine and was created with the vision to be the “Hippocampus for AI”. Unfortunately, the project went quiet after December 2024. KuzuDB, another embedded graph database with vector and full-text indexes similar to Cozo, was archived in October 2025.I decided to continue the vision by using Cozo to build a memory and reasoning substrate with a cognitive ontology inspired by neuroscience. It covers events, epistemic structure, intent, and action, with multi-agent support. An extraction pipeline using an LLM of your choice to extract these from documents and session transcripts, deduplicates them, and detects cross-document contradictions. Everything is versioned and provenanced back to source chunks, with time travel.But the features we rely on most needed work, so I hard-forked Cozo as MnesticDB under

parcle.ai• Jun 18, 2026• 1 min read

Show HN: We cut >60% of tokens from agentic tasks by removing repeated contextEvery agentic system I see has the same hidden tax: the model keeps rereading the same context.Tickets, Slack threads, docs, customer history, database notes, runbooks, logs, prior decisions. You can cache static prefixes, route to cheaper models, or set team budgets, but none of those fixes the underlying behavior: agents start most tasks trying to re-explore everything.We built Parcle as a shared memory layer for AI agents. It ingests operational context, indexes what happened, and lets agents retrieve a small, relevant memory set for the next step instead of pasting everything back into the prompt - or worse, letting the agent go explore on it's own and burning tokens.We started tracking our tokens consumed on tasks with and without our memory layer just with indexing of local files. In our deployments/evals, the biggest reduction we’ve seen is up to 70% lower token spend on agentic tasks, with roughly 2x faster task completion. The median was ~30% less tokens spent. The biggest sav

hydron.sh• Jun 4, 2026• 1 min read

Show HN: Hydron – Hardware-aware coding agentHi HN, this is Prashant from H2Loop. Embedded engineers that we work with were annoyed that generic AI tools hallucinated register addresses, generated code for peripherals that don't exist on the chip and mixed up timer quirks between similar platforms like STM32F4 and F7. The code looks clean but it just won't boot. This made them go back to the datasheet every time. So we built Hydron, an AI tool that writes datasheet-grounded code for your hardware.Demo: Hydron setting up sleep-mode CPU logging for an onboard temp sensor on an STM32U385 - https://boot.hydron.sh/zzzDemoFirst, we've pre-indexed 580+ platforms and peripherals. Most of what you'd use in a robotics, UAV, or IoT build: common dev platforms like STM32, ESP32, RP2040, AM6 families, plus the IMUs, GNSS modules, motor drivers, and baros that ship around them. Ask about a peripheral, the answer comes from our KG and the actual datasheet.Second, you can bring your own context and share it with your team. Hydron indexes PDFs up

github.com• May 21, 2026• 1 min read

Show HN: Darc – grep-like memory search tool for coding agentsHi HN, I’m Junha Park. I've been experimenting with agent memory, especially how to make agents run more reliably on large tasks.I built Darc, an open-source shared memory search tool for coding agents, with a different approach from most agent memory systems we can see today. It is an index + (lexical) search tool over agent session history, rather than a managed memory system. No embeddings, no agent-aided consolidation, no injection hooks.How it works: Darc archives Codex / Claude Code session rollouts that already exist under ~/.codex and ~/.claude, indexes them into a single SQLite DB, and exposes session/turn/tool call/file-level search commands over them.Why I built this: I've tried using different agent memory tools and they were useful, but sometimes I found them nosiy. I kept turning memory on/off depending on the task. For example, when I ran iterative code review rounds, I saw reviewer agents report "no findings" with "memory cited" memos, where the cited memory came from r

cbrincoveanu.github.io• May 6, 2026• 1 min read

Show HN: Gulugulu, an old-style client-side search engine for the old weird webA few days ago, I was looking for some obscure dev blogs and websites, and realized how completely useless normal search engines have become for finding "weird" content. Unless you already know the exact URL, almost every query just leads to SEO-optimized, commercial websites or AI-slop.So I built Gulugulu to fix this for myself. It's a search engine that only indexes the old/weird web like digital gardens, Neocities pages, ASCII art, and personal projects.There is no backend. It's a static site hosted on GitHub Pages.You can try it here: https://cbrincoveanu.github.io/gulugulu/The search runs entirely in your browser using Fuse.js against a single, flat index.json file. To get the data, I wrote a Python crawler that specifically scrapes curated indie webrings (like 512kb.club and Cloudhiker), extracts the basic metadata, and dumps it into the JSON array. Because it's completely serverless, there are zero analytics, no ads, and no cookie banners.Obviously, loading a massive JSON file i

toposonico.com• May 2, 2026• 1 min read

Show HN: A navigable map and recommender for 17M music entitiesHello HN,This is toposonico, a music recommender and navigable map. At core it's a skipgram word2vec model trained over ~6M playlists. Tracks are embedded in a 128d space. Embeddings for albums, artists and labels are computed marginalizing over tracks. The 2D map was built with UMAP.Both the model and UMAP were trained in the cloud over a NVIDIA A100. All things considered it cost me around ~50EUR, over two main training sessions and a few experiments. For the slippy map I experimented with a few libraries. Ended up with Maplibre GL JS. Loved working with it, kudos to their developers. For the recommender indexes I used FAISS, another fantastic piece of software. Pretty happy with the thing running on a small and cheap box.Two things influenced me in making this. The first is decade-old idea: human navigation and exploration skills work in information spaces too. Many ML concepts fit this idea especially well. It would be nice to see more experiments in this direction. The second goes

onboardly-ruddy.vercel.app• Apr 7, 2026• 1 min read

Show HN: Onboardly – Ask questions about any GitHub codebase in plain EnglishA few months ago, I asked HN about the biggest pain points when joining a new team (https://news.ycombinator.com/item?id=47368472). Most people said it was the tribal knowledge buried in the code.I've spent my time as a student building Onboardly to solve this. It’s a web app that indexes your GitHub repo and lets you ask questions in plain English. Unlike a general LLM, it cites the exact files it’s looking at so you can verify the logic.Tech stack: FastAPI backend, Next.js frontend, and a RAG architecture for the indexing. I’m currently hosting it on Render's free tier, so there might be a slight delay on the first login—I'm working on that!I’d love for you guys to break it and tell me where the answers feel 'off' or what features a dev actually needs for onboarding.It's free to try — would love feedback on answer quality across different repo types and languages.https://onboardly-ruddy.vercel.app

news.ycombinator.com• Mar 26, 2026• 1 min read

Show HN: Vizier – A physical design advisor for DuckDBHi,I've made an early version of a physical design advisor (called Vizier) for DuckDB. It can analyze a collection of queries (using a set of heuristics) and recommend changes to the physical design/layout of the database (for example, sort orders, Parquet layouts, indexes, etc.), in order to make those queries run faster.Vizier is implemented as a DuckDB extension in Zig and supports DuckDB version 1.2.0 and newer. The project is very early-stage, but if you're interested in learning more about Vizier or trying it out, you can check out the links below:Project's GitHub repo: https://github.com/CogitatorTech/vizierVizier documentation: https://cogitatortech.github.io/vizier/

news.ycombinator.com• Mar 14, 2026• 1 min read

Show HN: GitDB – GPU-accelerated vector database with Git-style version controlGitDB is a version-controlled vector database. Branch, merge, diff, and time-travel your embeddings the same way git handles source code. No server, no Docker — pip install gitdb-vectors and go. What makes it different: - Version control is native — git log, git diff, git branch, git merge for vectors. Roll back to any commit. Cherry-pick by meaning. - Time-travel queries — db.query_text("revenue", at="v1.0") searches an old snapshot. - CEPH CRUSH placement — deterministic data routing. Scales horizontally by adding peers, no coordinator. - P2P distributed — peers sync over SSH like git remotes. Each node is a full shard. - FoundationDB features — hooks, transactions, watches, secondary indexes, schema enforcement. - Universal ingest — swallow entire SQLite databases, MongoDB exports, CSV, Parquet, PDF with one command: db.ingest("legacy.db") - CLI-first — works from terminal like git: gitdb init && gitdb add --text "doc" && gitdb commit -m "init" - Embedded — no server process. Import

agentfork.dev• Mar 12, 2026• 1 min read

Show HN: AgentFork – Any repo, instantly runnable by AI agents or contributorsAgentFork lets maintainers and teams connect a GitHub repo. Our AI indexes the codebase — detects the framework, databases, services, and build steps — and generates an environment spec. Anyone can then fork the project and get a fully provisioned cloud environment with a live preview URL. Databases like Postgres and Redis are spun up automatically. The idea is that AI agents (or any contributor) can fork, make changes, and verify them against a real running instance without anyone configuring infrastructure. Early stage — waitlist is open at agentfork.dev. Would love technical feedback on the approach.

github.com• Feb 26, 2026• 1 min read

Show HN: Context Harness – Local first context engine for AI toolsContext Harness is a single Rust binary that gives AI tools like Cursor and Claude project-specific memory. It ingests docs, code, Jira tickets, Slack threads, and anything else into a local SQLite database, indexes them with FTS5 and optional vector embeddings, and exposes hybrid search via CLI and an MCP-compatible HTTP server.I built this because I kept hitting the same problem: AI tools are powerful but have no memory of my complex multi-repo project. They can't search our internal docs, past incidents, or architecture decisions. Cloud RAG services exist, but they're complex, expensive, and your data leaves your machine. I wanted something I could point at my sources and just run `ctx sync all`.Quick start: # Install (pre-built binaries available for macOS/Linux/Windows) cargo install --git https://github.com/parallax-labs/context-harness.git # Create config and initialize ctx init # Sync your data sources (filesystem, Git, S3, or Lua scripts) ctx sync all # Search from CLI ctx sea

type.lol• Feb 24, 2026• 1 min read

Show HN: Type.lol – Browse 800 independent type foundries, 14k typefacesI started type.lol in 2015 with a friend as a simple list of independent type foundries — basically a styled airtable doc. I'm a designer and I kept running into the same problem: I'd want to explore type beyond the usual distributors, end up with 30 tabs open, and lose track of what I'd already looked at. The list helped but it wasn't enough.I've since rebuilt it from scratch as a proper app. It now indexes 800+ foundries from 61 countries, 14k typefaces, and 1,500 designers. No signup required to browse. No ads, no sales, no commissions — it embeds actual foundry websites in iframes so traffic goes directly to them.You can browse by carousel, list, virtualized table, force-directed graph (showing relationships between foundries/designers/typefaces), or a 3D globe plotting foundries by location. Filter by classification, variable fonts, trial availability, country, designer, language support, or release era.Some technical details since this is HN:- React 19 + Vite + TypeScript + Zusta

jeescholar.com• Feb 20, 2026• 1 min read

Show HN: I indexed the academic papers buried in the DOJ Epstein FilesThe DOJ released ~3.5M pages of Epstein documents across 12 datasets. Buried in them are 207 academic papers and 14 books that nobody was really talking about. From what I understand these papers aren't usually freely accesible, but since they are public documents, now they are.I don't know, thought it was interesting to see what this dude was reading. You can check it out at jeescholar.com Pipeline: 1. Downloaded all 12 DOJ datasets + House Oversight Committee release 2. Heuristic pre-filter (abstract detection, DOI regex, citation block patterns, affiliation strings) to cut noise 3. LLM classifier to confirm and extract metadata 4. CrossRef and Semantic Scholar APIs for DOI matching, citation counts, abstracts 5. 87 of 207 papers got DOI matches; the rest are identified but not in major indexes Stack: FastAPI + SQLite (FTS5 for full-text search) + Cloudflare R2 for PDFs + nginx/Docker on Hetzner. The fields represented are genuinely iteresting: there's a cluster of child abuse/groomi

github.com• Feb 19, 2026• 1 min read

Show HN: Rememex – Semantic file search that runs 100% locally (Rust/Tauri)Hey HN, I built Rememex a semantic search layer for your local files. The problem: I kept losing files. Not because they were deleted, but because I couldn't remember the exact filename or keyword. grep needs the exact word. Everything only searches filenames. I wanted to type what I meant and find what I needed. How it works: - Indexes 120+ file types (code, docs, images, configs) - Hybrid search: vector embeddings + full-text + JINA cross-encoder reranking - OCR on images via Windows UWP engine - Reads EXIF GPS → reverse geocodes to city names ("photos from istanbul" works) - EXIF dates → human language ("summer morning" finds a July 8am photo) - Smart chunking per language (Rust at fn/struct, Python at def/class) - Built-in MCP server so AI agents can use it as a tool Everything runs locally. Embeddings use a local ONNX model (Multilingual-E5-Base) by default, though you can optionally plug in OpenAI/Gemini/Cohere. Named after Vannevar Bush's Memex (1945) his vision of a device that

clipthesis.com• Feb 19, 2026• 1 min read

Show HN: Clipthesis – free, local app to tag and search video across your drivesI'm a hobbyist video creator with a dumb problem: years of footage spread across multiple external drives with names like "C01456". Every time I sit down to edit, I waste too much time hunting for clips I know I shot but can't find.I tried the usual recommendations (NeoFinder, Lightroom, various DAM tools) but nothing fit how I actually work with video. The online alternatives seemed to charge an arm and a leg, which is hard to justify when you're not running a production studio.So I built a thing for a free Mac app that indexes all your drives into one searchable video library even when those drives aren't plugged in. What it actually does- Index any drive or folder — point it at your external drives and it catalogs every clip with thumbnails, metadata, duration, resolution, etc. - Tag + search — tag clips however you want (b-roll, drone, interview, whatever), then combine tags to find exactly what you need - Duplicate detection — content hashing finds the same clip across multiple dr