GitHub trends for September 25, 2026
What was taking off on GitHub on September 25, 2026: peak magnitude of the day, summary and where it was spotted.
-
8.8
paperclipai/paperclip
Open-source Node.js and React platform for orchestrating a team of AI agents: goals, org chart, budgets, tickets and audit. Suits teams running autonomous agents (Claude Code, Codex, Cursor, etc.) and wanting to manage…
-
8.7
vectorize-io/hindsight
Hindsight is a long-term memory system for AI agents that stores facts and observations rather than just conversation history, reporting state-of-the-art results on the LongMemEval benchmark. It runs as a server…
-
8.4
dream-num/univer
Open-source TypeScript SDK for embedding spreadsheet, document, and presentation editors into your own product, with a plugin architecture, Canvas rendering, a formula engine, and one Facade API for browser and…
-
8.3
google/ax
A declarative orchestrator for running autonomous AI agent workloads in a cluster, using a Kubernetes-like model with Task, Workspace, Gateway, and Model manifests. It sandboxes agents, pre-wires their environment, and…
-
7.8
mikehasa/golive-skill
Agent Skill and Node CLI that takes an agent-built app live: hosting, database, domain, email and payments on your own accounts. Runs a detect → plan → approve → apply → verify cycle with no hosted backend or telemetry.
-
7.7
driceroland/Search
A small, fast WebKit browser for macOS by design studio Office Commun: no toolbar, start page or accounts, with built-in ad blocking, reading mode and passwords kept in the keychain.
-
7.7
NandhaKishorM/laya
Laya is a non-autoregressive decision engine that answers typed questions (choice, score, noul) over any state (text, email, ticket, JSON) in a single forward pass (~35 ms on GPU) with calibrated probabilities.
-
7.6
Contrastive-LM/CLM
CLM is a class of System One language models trained contrastively to connect states and actions; the repo serves CLM-8B behind a TypeSafe-compatible API for fast agent decision-making.
-
7.5
tobi/disktree
A treemap utility for Omarchy built with Rust + GPUI: scans a disk, draws directories as tiles sized by real disk usage, and lets you mark clutter for deletion or trashing.
-
7.4
devdotfast/whiteboard
An open-source desktop IDE for software design where humans and coding agents share a canvas: the agent draws diagrams you can jump from straight into code.
-
7.2
strands-agents/harness-sdk
Open-source Python and TypeScript SDK for building and running AI agents: agent loop, tools, MCP, memory, multi-agent patterns and tracing in one package. Runs in your process with no hosted control plane and supports…
-
7.0
git-bug/git-bug
A distributed, offline-first bug tracker embedded in git: issues live inside the repository, work offline and sync through normal git remotes. It ships a CLI, terminal and web UI, plus bridges to GitHub, GitLab, Jira…
-
6.8
dgreenheck/tidewater
A browser fishing game set on a tropical island: cast from the pier, beach or boat, fight the fish, sell your catch and buy better gear. Runs on WebGPU with its own small rendering engine, no framework.
-
6.7
glaz-boga-poisk-bot/glaz-boga
Telegram bot for OSINT lookups: given a phone number, email, username, name or Telegram ID it returns matches from open, archived and leaked records. Matches depend on what is in the database.
-
6.7
JohnHeibel/PDoomVideo
Source code for a music video generated entirely by Claude Opus, with scenes, lyrics and transitions built in p5.js and rendered to MP4 via headless Chrome and ffmpeg.
-
6.6
jaredpalmer/kev
kev is a LoRA adapter with a small readout head on top of Qwen2.5-0.5B that answers many typed questions about a document in a single forward pass, returning calibrated probabilities instead of text.
-
6.6
jev-chat/jev-chat-jarvis
Android app that reads the on-screen chat via AccessibilityService, judges the other side's intent with the Jev model, and shows 3 ranked reply drafts in a translucent overlay. Replies are filled into the input box but…
-
6.5
derv82/wifit3
Cross-platform USB Wi-Fi auditor in Python: scans networks, captures WPA handshakes and PMKIDs, attacks WPS and WEP via its own userland drivers without aircrack-ng.
-
6.5
yetone/magpie
A menu-bar utility for macOS/Linux/Windows that lets you pick the model for each AI agent (Codex, Claude Code, Gemini CLI, etc.) and runs a local gateway translating OpenAI and Anthropic APIs between agents and…
-
6.4
nestrilabs/virtio-nvgpu
An experimental virtio device giving KVM guests near-native NVIDIA GPU access by forwarding driver ioctls, bypassing API-level translation. Aimed at headless streaming: a compositor in the VM renders and encodes frames…
-
6.3
amitshekhariitbhu/ai-system-design
A study guide to designing AI systems built on LLMs, RAG, and agents: from inference and GPUs to MCP, multi-agent systems, and interview preparation.
-
6.3
jamespolyakov9829/captcha-api
Dependency-free Python client and CLI for the Clearance captcha-solving API: one interface for Cloudflare Challenge and Turnstile with cost tracking and concurrent batch solves.
-
6.3
juspay/hyperswitch
Open-source composable payments infrastructure in Rust: routing transactions across 120+ providers, retries, a PCI-compliant vault, reconciliation and cost observability. For teams that want to own their payments stack…
-
6.3
Observal/Axl
A local-first agent harness for coding: a single daemon owns the session, model loop, tools, and history, while clients (TUI, SDK) connect over a typed protocol. For developers who want durable resumable sessions and…
-
6.3
ollaya-dev/ollaya
A local runtime for decision models: pulls and serves classification and routing models behind a TypeSafe-compatible API. Like Ollama, but for models that return probabilities instead of text.
-
6.2
Inkloom-art/inkloom
Specialised AI models for logo design: brand analysis turns a business into constraints, typography and symbol models construct the mark, and a composition engine produces lockups and clear-space rules. Generation is…
-
6.2
win4r/MuseAI-Skills
Unofficial archive of skill docs (SKILL.md) and a runtime snapshot of the muse.ai personal AI agent: 68 skills, connector manifests and permission lists. Meant for studying how triggers, task routing and deliverable…
-
6.1
ashleydarosa/CC-Checker-Validator-Generator
Python CLI toolkit for payment QA: generates synthetic Luhn-valid card numbers by brand, BIN, or pattern and validates them, including BIN/IIN lookup, expiry and CVV checks. Works offline, supports bulk validation and…
-
6.1
deepopen-com/deepopen
Open-source non-autoregressive System 1 decision engine for typed decisions: returns typed answers (choice, score, noul) in a single forward pass across 100+ languages without generating text. Suited for…
-
6.1
NVIDIA/Model-Optimizer
NVIDIA library for compressing and accelerating models: quantization, pruning, distillation, NAS and speculative decoding with export to TensorRT-LLM, vLLM, SGLang. For ML engineers preparing models for deployment.
-
6.0
Aureliengmz/clearwater
Real-time photoreal water demo in a single HTML file using WebGL2, no libraries or build step: FFT waves, caustics, refraction and glare.
-
6.0
hydra-db/hydradb
HydraDB is a distributed graph database written in Rust that stores the graph durably on S3-compatible object storage, with separate data nodes and indexers. It supports OpenCypher queries, GraphBLAS traversal, Neo4j…
-
6.0
rohitg00/ai-engineering-from-scratch
Open-source AI engineering curriculum: 523 lessons across 20 phases (~342 hours) in Python, TypeScript, Rust and Julia, where every lesson ships a reusable artifact such as a prompt, skill, agent or MCP server.
-
5.9
anthropics/financial-services
Reference agents, skills, and MCP connectors for financial-services workflows: investment banking, equity research, private equity, and wealth management. Installs as a Claude Cowork plugin or deploys via the Claude…
-
5.9
bridge-mind/bridgeclip
Open-source desktop app that turns long videos into short captioned clips: it finds the best moments, crops to 9:16 or 16:9, and burns in word-by-word captions. Runs locally, calling OpenRouter with your own keys.
-
5.9
harry7557558/spirula-studio
A self-contained C++ binary for training 3D Gaussian Splatting models: turns photos and video into splats and textured meshes, running on NVIDIA, AMD, Intel and Apple GPUs via Vulkan or CUDA without Python or COLMAP.
-
5.8
androoAGI/starnet
A local-first desktop app where AI agents do real work with real models and tools, and a pixel-art station projects live runtime state. Lets you create teams of agents with distinct permissions and run them concurrently.
-
5.8
ilkerzg/storycast
A web app that turns a topic into a narrated animated short film: script, keyframes, animation, voice-over, score and subtitles, all generated via fal APIs in the browser.
-
5.8
JohnHeibel/ClaudeAnimationBase
A starter kit for animating the hand-painted character Clawd in p5.js and p5.brush with Claude: code, 31 emotions, model sheets and the ANIMATION_GUIDE.md for the model.
-
5.8
samyost1/3dicon
A Claude Code skill that turns a text prompt into a looping animated 3D icon as an animated WebP with real transparency. Aimed at developers and designers who need icons for app UIs.
-
5.7
christianrowlands/wavedigger
Web app that looks up the approximate location of Wi-Fi access points by BSSID and LTE/5G NR cell towers by network parameters via Apple's public WLOC API, with map visualization.
-
5.7
togethercomputer/tev1
Open dataset and fine-tuning recipe for Qwen3.5-4B that picks one of 2–24 options given context and a question. The repo ships data, build scripts, Together AI training and evaluation tooling.
-
5.7
yu1745/wetype-ime-linux
Unofficial Linux port of the WeType (微信输入法) Android input engine with Fcitx 5 integration: the ARM64 engine runs under QEMU user mode, and the APK is downloaded from Tencent servers and patched locally.
-
5.6
alexgreensh/anidoodle
A tool for generating deterministic hand-drawn art and animation as code: nine styles, export to PNG, GIF, MP4, WebM and HTML. Works as an agent skill for Claude Code.
-
5.6
amitshekhariitbhu/ai-engineering-course
A free, structured AI Engineering course with 18 modules and 146+ lessons: from ML basics and Transformers to LLMs, RAG, agents, inference, evaluation, and AI safety.
-
5.6
angusdevgo/Seep-Reverse-Lab
Agent-native reverse engineering workbench for binaries and CWE-602 client-side authorization auditing: unifies Radare2, JADX, Apktool, Frida and IDA via 23 MCP tools with automatic task routing.
-
5.6
CopilotKit/openmuse
A personal AI agent with a browser, terminal, and files that carries out tasks on request and shows its work. Built with CopilotKit and AG-UI for iOS, Android, and web.
-
5.6
FxEmbed/FxEmbed
A service that fixes X/Twitter and Bluesky link embeds in Discord, Telegram and other messengers, adding support for videos, polls, translations and multiple images.
-
5.6
jankeesvw/omarchy-meeting-recorder
Records meetings on Omarchy: mic and system audio as separate tracks, transcribed locally with whisper.cpp, with speaker labels, chapters and a player.
-
5.6
kelseyhightower/kubernetes-the-hard-way
A step-by-step tutorial for manually bootstrapping Kubernetes without scripts or automation, to understand the cluster and its components. Aimed at engineers who want to learn Kubernetes fundamentals from the inside.
-
5.6
riba2534/claude-opus-5-5-demo
A demonstration of Claude Opus 5.5 code generation: three playable 3D web games, each built from a single one-sentence prompt in one session with zero human edits and deployed online.
-
5.5
852wa/JIZURA
A browser app that automatically assembles lyric videos (文字PV) from song lyrics using 707 built-in parts and 24 styles, exporting to MP4. An After Effects panel is also included.
-
5.5
eternity4719/HowToLiveBetter
An evidence-based life guide: 498 recommendations on longevity, first aid, personal finance, law, employment, family and emigration. Each lists costs, benefits, evidence level and primary sources.
-
5.5
leejet/stable-diffusion.cpp
Diffusion model inference (Stable Diffusion, FLUX, Wan, Qwen Image, etc.) implemented in pure C/C++ on top of ggml — like llama.cpp but for image and video generation. Runs without external dependencies on CPU, CUDA,…
-
5.5
mvt-project/mvt
A collection of mobile forensics utilities that gather and analyze traces of compromise on Android and iOS devices, matching them against public indicators of compromise (IOCs). Built by Amnesty International Security…
-
5.5
saragordic/rooms
A macOS app that groups windows into per-project "rooms": a hotkey restores a room's windows with a saved layout, hides the rest, and never closes anything.
-
5.5
sopersone/cabbage-trading-machine
Python spot trading bot built on the Investing Algorithm Framework: RSI/EMA signals, backtest, paper and live trading via CCXT.
-
5.4
codejunkie99/keel
A local-first macOS coding workspace built in Rust/GPUI: sessions, composer, terminal, coding-agent connections via ACP, and task-route selection by a local Laya selector or optional hosted Jev.
-
5.4
diggerhq/shipvideo
A web service that turns a URL or a text prompt into a 20–40 second launch video: an LLM writes an HTML animation and a serverless agent renders it to MP4 via headless Chromium and ffmpeg.
-
5.4
HKUDS/CLI-Anything
A collection of CLI harnesses that make desktop software controllable by AI agents via the command line, with a CLI-Hub registry for installation and updates.
-
5.4
LuwuDynamics/xgoduck_hardware
Hardware for the XGO Duck bipedal duck robot on Arduino Uno Q: printable body parts, expansion board, bill of materials, and assembly guide.
-
5.4
nokia-applied-research/AnyJev
Turns any LLM into a decision model with typed questions (choice, yes/no, score) that return real probabilities read from the next-token distribution, without training. L0 removes position and label-prior bias; L1 adds…
-
5.2
leter/zh-tech-writing
An Agent Skill for writing Chinese technical documentation based on Ruan Yifeng's style guide: short sentences, plain tone, no AI-speak. Works with Claude Code and other tools supporting Agent Skills.
-
5.2
openbao/openbao
OpenBao is an open-source secrets management system for storing, encrypting, and distributing credentials, keys, and certificates, with dynamic secrets and auditing. A community-governed fork of Vault.
-
5.2
unreallabsai/unreal-agent
Async-first agent harness in Go: sessions with persisted forkable history, input deduplication, an LLM-turn coordinator, and asynchronous operation execution via swappable components.
-
5.1
farion1231/cc-switch
Cross-platform desktop manager for Claude Code, Codex, Gemini CLI, OpenCode and other agentic tools: configure and switch between providers and models in one place.
-
5.1
kaolti/phantomat
A Hyprland plugin that replaces workspaces with an infinite 2D canvas: windows float freely on one plane, and zooming plus title search let you move between them quickly.
-
5.0
cdyforever/how-to-live-better
A single-file HTML reading page for the open book "High Cost-Performance Life Guide" — 528 tips across 32 sections, searchable, offline-capable, zero dependencies.
-
5.0
dominique1993/Crypto-Whale-Tracker
Real-time Ethereum whale tracker: streams pending transactions over WebSocket, decodes ETH/ERC-20, labels exchange wallets and sends Telegram alerts when transfers exceed a configurable USD threshold. Data is stored in…
-
5.0
Human-Agent-Society/reef
Continual learning infrastructure for self-improving agents: connects agent inference, feedback collection, learning, and versioned delivery of updates. Lets you train model weights or improve an agent's harness…
-
5.0
Liuziyu77/Valen
A multimodal decision model built on a Qwen3.5-0.8B/2B backbone: it takes text, images and video with an instruction and returns probabilities over supplied candidates without generating answer tokens. The repo…
-
5.0
mvschwarz/openrig
OpenRig is a harness orchestrator for AI coding agents: a team of agents (Claude Code and Codex) is defined in YAML and booted with one command, turning scattered terminal sessions into a managed team with a lead agent.
-
5.0
nobodywho-ooo/nobodywho
A Rust inference engine for running LLMs locally with bindings for Python, Kotlin, Swift, Flutter, React Native and Godot, plus speech-to-text and text-to-speech. It runs GGUF chat models offline on desktop and mobile…
-
5.0
robbietilton/Compositor
A free, open-source raster image editor for macOS written in Swift — a Photoshop alternative focused on compositing and post-processing.
-
4.9
BuilderIO/agent-native
An open-source TypeScript framework for building agentic apps where each capability is defined once as an action, used by the agent as a tool and called by the UI from code. Aimed at developers who need agents paired…
-
4.9
lukeTheNeuromancer/system-design-notes-zh
Chinese translation of notes based on Alex Xu's "System Design Interview" books (Volumes 1 and 2), covering 28 chapters from scaling to designing a stock exchange. Material for preparing for system design interviews.
-
4.9
mizorewww/laya-mlx
Native MLX runtime for Laya typed decision models: returns probabilities for choices, scores or truth values without text generation, running locally on Apple Silicon.
-
4.9
sunkeke-ai/Life-Compass-Obsidian-CN
Unofficial Simplified Chinese localization of Compass, a life-management system for Obsidian covering journaling, goals, habits, tasks, projects and reviews. Keeps the original Markdown structure and workflow…
-
4.9
TNT-Likely/PanWatch
Self-hosted AI stock-watching assistant for A-shares, HK and US markets: real-time quotes, portfolio tracking, technical analysis and IM notifications. Integrates TradingAgents for multi-agent position analysis with…
-
4.8
JaxsonWang/cpa-plugin-oai-basispoints
A native CLIProxyAPI plugin that exposes models like gpt-6-astra-basispoints over the Responses protocol using existing ChatGPT/Codex OAuth credentials.
-
4.8
julyx10/lap
Open-source local-first photo manager for macOS, Windows, and Linux that works directly with existing folders, no cloud or uploads, with local AI search and face recognition.
-
4.8
qybaihe/mu
A coding agent with a judgment kernel: a small, fast judge model makes routine calls (command risk, what stays in context, when work is done) while the big model does the coding. Ships as a CLI and a desktop app.
-
4.8
stablyai/orca
Orca is an ADE for running a fleet of parallel coding agents (Claude Code, Codex, OpenCode, etc.) in isolated git worktrees with a unified interface, mobile companion, and remote runtime.
-
4.8
superdesigndev/treg
Treg is a proxy registry of tools for AI agents: one base URL and token give access to 3,000+ endpoints across 60+ providers (SEO, social, data enrichment, media generation) with per-call pricing and no provider signup.
-
4.8
TeamGDB/Yakumo
A native port of Monster Hunter Portable 3rd HD Ver. via static recompilation of PSP code into C++, rendering with Vulkan and SDL3; not an emulator, requires your own game copy.
-
4.8
tjsky/pinyin-annotator
A web tool that adds pinyin annotations to Chinese texts, producing reading materials for children from kindergarten to early primary school. Runs entirely in the browser without uploading text to any server.
-
4.7
anishfn/shapeshift
A web app with a single text box that morphs into the right UI as you type: event card, checklist, timer, color picker and more. Intent is classified by the Jev model, while dates and numbers are parsed…
-
4.7
Avinash-jetwani/jevmem
Automatic project memory for Claude Code, Cursor and Codex: it saves decisions, constraints, bugs and todos from chats into JEVMEM.md and injects relevant lines into the next session's context.
-
4.7
Badtheorylabs/interference-search
A search method that reasons over explicit states instead of a language model's linear transcript: many branches expand at once, duplicates merge, dead ends are dropped by a trained judge, and survivors advance…
-
4.7
browser-use/video-use
A skill that lets coding agents (Claude Code, Codex) edit raw video footage via chat: it transcribes takes with ElevenLabs, builds an edit decision list, and renders final.mp4 with cuts, color grades, subtitles and…
-
4.7
davila7/claude-code-templates
CLI tool for installing and configuring ready-made Claude Code setups: agents, slash commands, MCP integrations, hooks and settings, plus session monitoring. Aimed at developers using Claude Code.
-
4.7
deepseek-ai/deepseek-harness
Open-source agent harness by DeepSeek AI written in TypeScript, built on an everything-is-a-plugin architecture powered by Cordis. It runs as a web UI via npx and serves as a base for composing AI agents from plugins.
-
4.7
dsummersl/treepeat
A Python tool for finding duplicate, near-duplicate, and structurally similar code using tree-sitter and locality-sensitive hashing. Supports multiple languages and configurable normalization rulesets.
-
4.7
mike007jd/voxel-musou
A browser-playable voxel action game in the style of Dynasty Warriors built with Three.js: play as Zhao Yun and cut through hundreds of Wei soldiers with his spear. No build step — plain ES modules and a fixed 60 Hz…
-
4.7
op7418/guizang-product-video-skill
A skill for AI coding agents (Claude Code, Codex) that turns a product's codebase into a release promo video: reuses real components, writes the script, generates music and SFX, and renders an MP4.
-
4.6
blixvip/NullMotion
A web tool that plays a finished motion-graphics ad over its black-and-white HyperFrames drafts, keeping them in sync with the film and letting you export everything to MP4 frame by frame.
-
4.6
browser-use/jev-ultrafast
A browser agent with a dynamic, indexed action space: given one natural-language goal it picks an operation (CLICK, TYPE_TEXT, SELECT, etc.) and an element, using a small LLM to generate text only for typing.
-
4.6
chelinho139/wow-ai
A World of Warcraft addon that lets you chat with local coding agents (Claude Code, Codex, Grok) from inside the game via a Node.js bridge, without alt-tabbing.
-
4.6
Kaixxrua/excel-codex-bridge
A local bridge that lets Codex CLI and Codex Desktop run through an existing ChatGPT login on your machine (from Codex or the Excel add-in cache), without implementing its own OAuth.
-
4.6
michaelswissa/jevry
An MIT-licensed Electron desktop browser that turns natural-language tasks into structured actions: the Jev model picks the next action from the page's actual capabilities, while text and vision models handle planning…