Trending archive for Oct 7, 2026

openai/math

OpenAI's model-generated math proofs, with Lean formalizations in progress — A collection of 719 mathematical manuscripts in 372 families produced by an internal OpenAI model tackling open research problems, released as model-development evaluations saturated on existing benchmarks. About 42% of top-line results have Lean formal proofs so far, with reasoning summaries and a comparator tool included for verification.

Niko1221/Strata

A 125B-parameter model, squeezed onto your gaming PC — Strata is a one-click Windows/Linux installer that runs Qwen3.8-Flash-Next, a 125-billion-parameter model, locally on consumer NVIDIA, AMD or Intel GPUs. It exposes an OpenAI/Anthropic-compatible API on localhost, supports image input, and keeps chats and files on-device, useful for coding agents and local app integrations without cloud dependency.

Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on localhost, optional image input.

omlahore/RemoveMacAI

A reversible, GUI-friendly way to strip Apple Intelligence from macOS — RemoveMacAI offers a native app and CLI to disable Apple Intelligence, delete its models, stop analytics and ads, and quiet pop-ups on macOS, reclaiming disk space. Every change is previewed and reversible. It builds on pared's research into Apple's asset service and settings, and has been covered by The Verge, Ars Technica and The Register.

Debloat macOS: turn off Apple Intelligence, analytics, ads and pop-ups. A native app and CLI, and every change can be undone.

streetcomplete/StreetComplete

Quest-based OSM editing now spans Android and iOS via KMP — StreetComplete turns OpenStreetMap contribution into simple on-site quests like naming a road or describing its surface, no tagging knowledge required. Built as a Kotlin Multiplatform app, it now targets both Android and iOS, works offline, and uploads answers directly as OSM changesets.

Easy to use OpenStreetMap editor for Android and iOS

szabadkai/c64-keyboard-font

A font that recreates the actual lettering on C64 keycaps — Built from photographs of a real Central European Commodore 64 keyboard, this font reproduces the printed key legends, including all 63 PETSCII graphic symbols and function-key labels, as clean geometric outlines rather than the screen bitmap font. Useful for retro-computing designers, emulator UIs, or print projects needing authentic C64 keycap typography.

Photo-based reconstruction of classic Commodore 64 keycap lettering, with desktop fonts, webfont, editable outlines, and a specimen.

nanaism/yomiyasu

An Agent Skill that cleans up AI-sounding Japanese prose — A Claude Code/Codex/Cursor skill that fixes the structural tics of AI-generated Japanese—vague metaphors, unclear subject-predicate links, inanimate subjects—rather than just banning buzzwords. Aimed at writers of technical docs, specs, PR descriptions, and internal reports who need naturally readable Japanese.

AI生成の日本語を自然な日本語へ推敲するAgent Skill / Agent Skill for Refining AI-Generated Japanese into Natural Japanese

coji/natural-japanese

An Agent Skill that strips AI stiffness out of Japanese work writing — This Claude/Codex Agent Skill rewrites AI-generated Japanese reports, minutes, and PR descriptions into natural phrasing without adding facts or changing certainty levels. Tests across 25 documents show bold-text overuse dropping from 551 to 13 instances, aimed at writers polishing AI drafts.

仕事の日本語を、読みやすくわかりやすく書く・直すための Agent Skill です。

cursor/plugins

Official spec and plugin registry for extending Cursor agents — Defines the manifest format for Cursor plugins and ships a growing collection of official ones, from CI workflows and PR review canvases to self-referential AI loops and repo compatibility scanners. Useful for developers who want to extend Cursor's agent behavior or build their own plugins.

Cursor plugin specification and official plugins

michael-denyer/pstack-claude

Porting Cursor's pstack skill stack to every major coding agent — This repo ports Lauren Tan's pstack, an opinionated Cursor skill stack, to Claude Code, Codex, Pi, Copilot, Gemini and others. It tracks upstream changes plus named policy forks, and adds a formal-verification skill for concurrency bugs tests can't catch. Useful for anyone running multiple agent harnesses who wants consistent, verified workflows.

Claude Code, Codex, Copilot, Pi, OpenCode, Muse, Gemini, and Prime Agent versions of Poteto's pstack. Rigorous agent workflows with Cursor primitives translated for other harnesses.

microsoft/graphrag

Microsoft's graph-based RAG pipeline, now in maintenance mode — GraphRAG turns unstructured text into structured graph data to give LLMs targeted context for question answering over private data. Microsoft Research built it as a research exploration; it's now largely maintained for bug fixes and CVEs rather than new features. Useful for those wanting a documented, citable RAG architecture rather than active new development.

A modular graph-based Retrieval-Augmented Generation (RAG) system