GitHub trends for September 27, 2026
What was taking off on GitHub on September 27, 2026: peak magnitude of the day, summary and where it was spotted.
-
9.6
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…
-
9.4
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…
-
7.8
mexicat/pdoom-video
A generative, code-rendered music video built with TypeScript and three.js, where every scene is synced to the song's lyrics and each frame is a deterministic function of track time. It includes a Python audio-analysis…
-
7.5
dzhng/jevgrep
A CLI for coding agents that finds relevant files and source excerpts in a repository from a question about code behavior. It uses Jev for relevance ranking and requires a Vercel AI Gateway, TypeSafe, or OpenRouter key.
-
7.4
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…
-
7.4
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…
-
7.3
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.
-
7.2
reladraw/reladraw
A text language for diagrams where element placement is stated in words (above-left of, between) instead of coordinates; a TypeScript parser, resolver and SVG renderer plus a CLI.
-
7.0
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…
-
7.0
supermemoryai/company-brain
Open-source version of a company brain for Slack: it remembers team conversations, answers from them, and performs tasks in connected tools (GitHub, Linear, Notion, etc.). Runs on your own Cloudflare account.
-
6.9
Niko1221/Strata
A C++ local inference engine that runs the 125B MoE model Qwen3.8-Flash-Next on a regular PC with one NVIDIA GPU (12-24 GB) and 64 GB RAM, exposing an OpenAI/Anthropic-compatible API on localhost.
-
6.9
yihui-dev/awesome-opus5-5-videos
A collection of 282 viral videos made with Claude Opus 5.5, each with the creator's exact prompt. The prompts generate animations as code using HTML, Canvas, SVG, Three.js and GLSL.
-
6.8
InfinityLoop1308/PipePipe
A NewPipe fork for Android that lets you browse YouTube and other video services without ads, with SponsorBlock, ReturnYouTubeDislike, background playback and playlist downloads.
-
6.6
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.
-
6.5
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.
-
6.5
feitangyuan/onetake
A Claude Agent Skill for making product launch films and feature demos as one continuous take, where each beat grows out of the previous one and continuity is verified by an oracle.
-
6.5
xikhar/spiderbench
A browser-based 3D web-swinging game built by Claude under human direction as a non-commercial fan project and benchmark of AI game-development capabilities.
-
6.4
lemomo-ai/lemo-opuscar
A collection of 39 film styles, each with a reusable style prompt, a short film made entirely in code (Canvas/WebGL, TTS, ffmpeg), and DIRECTOR.md/TECHNIQUE.md guides for an AI agent director.
-
6.4
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.
-
6.4
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.
-
6.3
debpalash/VoiceStudio
A fully local, open-source ElevenLabs alternative for voice cloning, voice design, video dubbing, dictation, transcription and audiobook creation on your own hardware, with no account or API key. Supports 16 TTS and 11…
-
6.3
vercel/next.js
Next.js is a React framework for building full-stack web applications with server rendering and Rust-based build tooling. Used by large companies for fast, universal sites.
-
6.2
Soniavasseur/Wallet-Risk-Scanner
Multi-chain crypto wallet risk scanner for AML/KYT screening: scores addresses 0–100 using sanctions lists, risky contracts and fund tracing across 6 intelligence providers and 15+ blockchains.
-
6.1
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.
-
6.1
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.
-
6.1
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.
-
6.0
anthropics/claude-code-action
A GitHub Action that runs Claude Code on PRs and issues: answers questions, reviews code, and implements changes triggered by @claude mentions or explicit prompts. Supports Anthropic API, Bedrock, Vertex AI, and…
-
6.0
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.
-
6.0
shengjidaguai-china/goutoujunshi
A Codex Skill for relationship analysis: it first offers emotional support, then separates facts from guesses and gives concrete next steps. It draws on a knowledge base covering psychology, law, sociology and…
-
5.9
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.
-
5.9
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.
-
5.9
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.
-
5.9
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…
-
5.9
HeyOkay/HaloBattery
Windows utility that shows battery levels of wireless mice, headsets and Bluetooth devices as separate tray icons, without vendor software.
-
5.9
Rieranthony/product-film-skill
A Claude Code skill that produces showreel-grade product films with Remotion: it learns your design system, writes the story, cuts music on the beat and renders the final film.
-
5.8
athemeroy/awesome-claude-5-5-videos
A curated, source-linked guide to 1,000+ videos made with Claude Opus 5.5, covering topics, visual styles, and how the frames were produced.
-
5.8
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…
-
5.8
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…
-
5.7
glanderness/BeefTV
A local-first desktop video workspace with AI agents: on a free canvas it links models, assets and generation tasks for text, images, video and audio into an editable workflow.
-
5.7
latent-spaces/brag
An agent skill (Claude Code, Codex, opencode, etc.) that turns a project you just built into a short launch video with music, motion, and share copy — in one command. Rendering is powered by Hyperframes.
-
5.7
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.
-
5.6
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.
-
5.5
cclank/lanshu-create-ai-presenter-video
Codex Skill for producing AI presenter videos: from a topic or script and an authorized person image it assembles voiceover, character generation, lip sync, subtitles and editing with QA.
-
5.5
yasinozmeen/animasyon-stil-katalogu
A catalog of 20 animation styles for videos: each style is demonstrated by its own 9–10 second clip, selectable from the README before producing a new video.
-
5.4
AgentSystemLabs/agent-office
A 3D cartoon office where a team hires Claude Code, OpenCode or Codex workers at desks, watches their live terminals, chats by voice, and tracks GitHub issues and PRs per project floor.
-
5.4
mobile-next/mobile-mcp
MCP server for mobile automation on iOS and Android: agents and LLMs drive emulators, simulators and real devices via the accessibility tree or coordinate-based taps.
-
5.4
MysticGSI/mysticgsi
A tool that builds a Generic System Image (GSI) from stock Android firmware of many vendors: OTA zips, fastboot packages, super.img, Samsung tars, Huawei UPDATE.APP and more.
-
5.3
actions/runner-images
Source code for VM images of GitHub-hosted runners in GitHub Actions and Microsoft-hosted agents in Azure Pipelines: Ubuntu, Windows, macOS with preinstalled software.
-
5.3
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.
-
5.3
deepfates/imp
A port of DSPy to the BEAM: declarative LM programs in Elixir with signatures, optimizers and agents running as OTP processes.
-
5.3
Themaister/pyrowave
PyroWave is an intra-only video codec (essentially a still-image codec) implemented in Vulkan compute shaders, optimized for extremely fast GPU encode/decode (<0.1 ms at 1080p). It targets local network game streaming…
-
5.2
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.
-
5.2
markxplorer969/profile
Repository containing only a README file with the heading "profile". No code, description, or purpose is provided in the README.
-
5.2
Roberto02800/turnstile-token
A dependency-free Python client and CLI for obtaining Cloudflare Turnstile tokens via the paid Clearance API: it discovers the sitekey on a page, solves the challenge, and returns the token together with the browser…
-
5.1
Barty-Bart/motion-graphics
A set of motion-graphics skills for Claude Code and Codex. The first skill, motion-broll, takes a video and transcript and plans, animates and renders B-roll clips timed to your words.
-
5.1
bjarneo/flux
Connects an Omarchy desktop and an Android phone over the local network: file and clipboard sharing, notifications, media, camera and microphone. Includes a CLI, daemon, Qt window, shell plugin and Android app.
-
5.1
brumar/chess-postmortem-skills
A set of Claude Code skills that turn a chess game into a post-mortem: Stockfish analysis of every move, annotated PGN, HTML board, and a narrated video. It can also use a recording of the player's think-aloud during…
-
5.1
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…
-
5.0
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.
-
5.0
amirmushichge/unreal-home-wizard
A guided Codex workflow that turns photographs of a room or house into a reviewable, walkable Unreal Engine project. Aimed at complete beginners and does not promise a measured digital twin from photos alone.
-
5.0
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.
-
5.0
kaankiziltug/logo-design-skill
A skill set and tools for Claude that turns it into a disciplined logo designer: principles, process, SVG craft, testing tools and a 1,400+ reference library.
-
5.0
SinaXhpm/Submarine
Cross-platform SSH/SFTP client with terminal, file browser, port forwarding, folder mirroring, and encrypted profile sync. Replaces the PuTTY + WinSCP + tunnel manager stack with a single app.
-
5.0
SunnyMaria/csapp-zh-markdown
A Chinese Markdown edition of the CSAPP third-edition textbook: all chapters, 238 exercise answers, and translations of eight labs, for online or offline study.
-
5.0
tigerless-labs/autoharness
A Claude Code plugin that distills skills from your real sessions, merges similar ones, updates those in use, and prunes unused ones. Runs as a Python hook and MCP server with no third-party dependencies.
-
4.9
avinash201199/System-Design-KIT
A curated collection of resources, case studies, tools, and roadmaps for system design — from basic APIs to distributed architectures. For interview prep and engineer growth.
-
4.9
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.
-
4.9
imikerussell/beebots
An experimental platform for automated crypto perpetuals trading on OKX: three AI "bees" make decisions via the Jev model, with orders passing through a code-based risk layer; runs in paper trading mode by default.
-
4.9
ScottieFox/caustic-volume
Real-time water caustics in the browser built on three.js: sunlight bends through a moving water surface and gathers into bright shifting lines on the pool floor. Two single-file HTML versions, lite and sandbox, with…
-
4.9
tswawa/WechatVibe
Desktop client for analyzing emotions and intents in WeChat chats: recognizes intents, emotions, builds profiles of contacts and groups, scores affinity and infers MBTI. Runs locally on the Laya ONNX model or via a…
-
4.8
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.
-
4.8
Ebony-Vinyl/dsh-our-free-model
A dsh plugin that adds free access to Muse Spark 1.3, MiMo V2.6 and other models without sign-up or API key, plus a local OpenAI-compatible proxy and usage dashboard.
-
4.8
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…
-
4.8
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.
-
4.7
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.
-
4.7
Mapika/decider
A language model that does not generate text: from a single forward pass it returns calibrated probabilities for typed questions (Choice, Score, Noul) about a given state. An open reproduction of the "System One" model…
-
4.7
microsoft/vscode
Open-source repository of Microsoft's Visual Studio Code: a cross-platform code editor with editing, navigation, debugging and extensibility. It is the base for building Code - OSS and for community development.
-
4.7
zhaoxuya520/reverse-skill
A skill-router pack that guides AI coding agents through reverse engineering, pentesting, and security tasks (APK, binaries, JS, CTF) to the right methodology and toolchain instead of guessing commands.
-
4.6
InternLM/Intern-Decision
A multimodal decision model built on Qwen3.5 that turns a state, optional images, and typed questions (choice, score, yes/no) into decisions with probabilities. Ships with training, two inference backends, temperature…
-
4.6
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…
-
4.6
opusvideo/awesome-claude-video
A curated collection of videos and animations made with Claude Opus 5.5: demos, original posts, prompts and workflows with full implementation guides.
-
4.6
Rizzo-AI-Academy/rizzo-flow
A local-first implementation of the Jev idea: an LLM returns typed decisions (boolean, choice, score, numeric) with probabilities in a single forward pass, without generating tokens. Compatible with the TypeSafe HTTP…
-
4.6
sciaschi/CBFD-Recompiled
A native PC port of Conker's Bad Fur Day (N64), built by static recompilation with N64Recomp, running on N64ModernRuntime and rendered by RT64. Requires your own legally obtained US ROM.
-
4.6
tigerless-labs/agent-memory
Local long-term memory runtime for AI agents: Markdown files as the source of truth, ranked retrieval over a SQLite index, and sleep-time consolidation. One shared store for Claude Code and Codex, no API keys.
-
4.6
tugrawork-creator/saas-motion-kit
A toolkit and methodology for producing promo and motion videos for software products with HyperFrames and Claude Code, focused on keeping effects and transitions varied across films.
-
4.6
YUTA-fywoo/jev-gui-delegate
A GUI delegation prototype for Codex on Windows and Chrome: a local controller runs deterministic steps, calls Jev for semantic choices, and verifies outcomes.
-
4.6
ZJU-REAL/Easel
An open-source AI agent for social media: it discovers trends, creates content, publishes it to Xiaohongshu, Douyin, Zhihu, Bilibili and more, and analyzes what performs. Aimed at content creators and social media…
-
4.5
christianmat/jev-pokemon
An experiment where the AI model Jev plays Pokémon Red through a Game Boy emulator: the harness reads game memory and feeds the model legal options, while all decisions are made by the model.
-
4.5
DanFessler/trellis
A layout engine for web apps: panels nest to any depth, the view zooms to the part you need, and every view stays mounted. Ships as a framework-agnostic core, a React adapter and a web component.
-
4.5
pacifio/atlas
Atlas is source control for coding agents: it links every commit back to the agent session with its prompts, tool calls and reasoning, and lets you run multiple agents side by side with shared memory.
-
4.4
admte/orc-claude-plugin
A Claude plugin that connects Claude to the ORC8R Cloud MCP server via OAuth, with a skill for finding operations, inspecting their inputs, and reporting results.
-
4.4
aws-samples/sample-voice-ai-agent-studio
A studio for designing voice AI agents on AWS: a wizard to configure model, tools, RAG, and channel, deployed on Amazon Bedrock AgentCore. Built on Amazon Nova 2 Sonic and Strands BidiAgent.
-
4.4
Jiarui-0410/multimodal-product-discovery
Full-stack multimodal product retrieval and personalized reranking: CLIP and FAISS fetch candidates, while Spring Boot and PostgreSQL store profiles and rerank results.
-
4.4
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.
-
4.4
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…
-
4.4
Mak5er/AirCard-iOS
iOS app that writes custom artwork into iOS 27 system caches over a local loopback tunnel, letting you restyle Apple Wallet cards and the lock screen passcode keypad directly on the iPhone.
-
4.4
pingwin010/tiktok-views-bot
A Python tool for the TikTok private API: generates X-Gorgon, X-Khronos, X-Argus, X-Ladon, TTEncrypt signatures and handles device registration. Used for view boosting and automating TikTok requests.
-
4.4
putyy/res-downloader
Cross-platform resource downloader built with Go + Wails: it sniffs traffic through a proxy and downloads videos, audio, images, m3u8 and live streams from WeChat, Douyin, Kuaishou, Xiaohongshu, Kugou and QQ Music.
-
4.4
sopersone/cabbage-trading-machine
Python spot trading bot built on the Investing Algorithm Framework: RSI/EMA signals, backtest, paper and live trading via CCXT.
-
4.4
tensorflow/tensorflow
An open source machine learning platform from Google Brain with Python and C++ APIs for training and deploying neural networks. Used by researchers and developers of ML applications.