GitHub Tools
1279 best GitHub tools and apps, curated and ranked by community upvotes on ProductCool. Updated daily as new GitHub products launch.
Ruby UTCP brings UTCP 1.1 to Ruby, giving apps and AI agents a standard way to discover and call tools over native protocols. It supports 12 transports, including HTTP, CLI, WebSocket, gRPC, GraphQL, MCP and WebRTC, plus streaming, auth, OpenAPI discovery, and CodeMode for programmable multi-tool workflows. Open source and built for the Ruby ecosystem.
Forma turns a prompt or image into an editable hardware project with parts, wiring, mechanical layouts, cost estimates, and assembly instructions. Its open-source, local-first workspace connects supported AI agents to shared project data, rule-based electrical checks, and CAD tools. Generate concept renders, inspect the design, and revise your project as your idea evolves.
PC Anatomy is an open source interactive 3D explorer for computer hardware. Take apart a desktop PC, inspect its components, and dive deeper from the motherboard, CPU, cooling, storage and GPU all the way down to GPU architecture and individual streaming multiprocessors. Every component includes explanations, specifications and sources, with all 3D geometry generated directly in TypeScript and Three.js.
GitDiagram is a free tool that instantly turns any GitHub repository into an interactive visual diagram. It solves the problem of navigating complex codebases by providing a clear, visual map of the repository's structure and dependencies. It's designed for developers, technical leads, and open-source contributors who want to quickly understand, explore, or present a project's architecture without cloning or manually tracing the code.
Run GPU jobs from multiple Codex tasks through one local Windows queue. A shared broker starts eligible work automatically. Inspect confirmed waiting blockers in a redacted, read-only CLI; uncertain reasons stay undetermined.
moli: 40MB headless browser for AI agents. Lite, Fast, High-Compatibility. Built in Rust
Loci is a free, open-source desktop workspace for scientific images. View microscopy and 3D data, count cells, annotate, measure and export traceable results without a cloud account or subscription. Use built-in analysis or bring compatible model packages. Built by a researcher and shaped by real lab workflows. Currently public beta for Apple Silicon Macs; research use only.
Turn almost any Bluetooth remote into a driverless keyboard and media controller for your PC with a $5 ESP32-C3. No apps, drivers, admin rights, or system modifications. Flash, configure Wi-Fi, and remap every key directly from the browser. Supports Xiaomi RC003 and standard BLE remotes, with live key feedback and fully local processing.
M9R is the multiplayer layer for AI coding agents. Bring Claude Code, Codex, OpenCode, and other supported agents into one shared workspace. Let them communicate across providers, hand work off, and let teammates join the same work instead of juggling isolated tabs. Not another model. A better place where agents work together and talk to everyone in the space.
Holon is a local workbench for multiple agents with ongoing responsibilities—not just one-off chats. Agents handle development, review, documentation, and operations, saving progress and resuming when tests finish, feedback arrives, or you make a decision. Check in through a terminal or browser without starting over.
Deploy your own private scanner on ChatGPT Sites. Capture receipts hands-free with your phone, then use ChatGPT Work to parse receipts and invoices. Stress-tested with 800 receipts so far and continuously improved through real-world use.
Claude Code already logs every token it bills. claudebill reads those logs and shows cost per session, project, git branch, model, and day, plus what prompt caching saved you. Retroactive, offline, zero dependencies. npx claudebill.
AI agents make CI pass, but a green check does not prove the failure was fixed. Sutura is a GitHub Action and CLI that reproduces the failure in an isolated sandbox, separates flakes from real failures, searches bounded repairs, rejects green-wash (deleted tests, weakened assertions, relaxed config), then puts the winner through an adversarial audit by NVIDIA Nemotron with GPT-6 Astra and TypeSafe Jev as veto-only second opinions. It opens an evidence-backed PR for a human. Never auto-merges.
Wombo allows you to easily create sprites and sound assets for your game. It implements chroma keying so you get a sprite animation ready-to-use, but you also have native Aseprite exports ready for editing. It generates aligned side/front/back views so you can create animation sequences from any angle. Under the hood it combines GPT-Image-2.5, video generation (MiniMax H3) and ffmpeg to extract, resize and align the frames.
omadesign is a native Linux studio for design, layout, paint, photo, and motion. One document, one layer stack. Open-source (MIT). Built for Linux designers (including Asahi / Omarchy): not GTK or Electron. Install into ~/.local and start a file. v0.5.0: Design, Layout, Pixel, Photo, and Motion with templates, brand kits, and layered file interchange. Still alpha; rough edges welcome as bug reports. Install: curl -fsSL https://omadesign.app/install | sh Site: https://omadesign.app
Forma turns a prompt or image into an editable hardware project with parts, wiring, mechanical layouts, cost estimates, and assembly instructions. Its open-source, local-first workspace connects supported AI agents to shared project data, rule-based electrical checks, and CAD tools. Generate concept renders, inspect the design, and revise your project as your idea evolves.
AtlasAlign is an open-source Fiji/ImageJ2 plugin for mouse brain atlas alignment and ROI export. Beta.3 adds multichannel and composite viewing, registration pinned to C/Z/T, selected-plane exports, and saved review sessions. Refine alignment, edit anatomical ROIs, and export crops and masks in original image coordinates. Resume individual or batch reviews. Source images stay read-only; researchers review and accept the anatomy. Demo frames show the earlier interface.
Sim is the open-source AI workspace where companies build, deploy monitor, and govern agents and workflows. Now with Sim Search, you can turn your personal and company data into a knowledge graph.
Powermove is an iterative video editor that rewrites itself around your work. Ask an agent to add or fork panels, effects, and workflows, then load them into the running app. If it is wrong, steer it and try again. Every change can be undone.
Keeps Chrome tabs healthy: open a duplicate and it switches you to the tab you already had, untouched tabs fold into a collapsed Stale group, right-click → Snooze brings a tab back later today, tomorrow or any date. Undo anything. Offline, open source.
Second Eyes helps you sort photos, understand the shortlist, and make the final call. For business briefs or personal needs, see why each photo was shortlisted and what still needs checking. You decide what to publish. Explore a real media editorial project involving 404 source files, then try six public sample photos to see how your requirements reshape the shortlist. This demo uses a recorded review. No uploads yet. Would you use this with your own photos? Share your workflow in the comments.