A framework of portable agent skills that prevents AI from generating generic boilerplate UIs, enabling the creation of premium frontends with professional design and motion.

An open-source factory for one-shot engineering that automates routine development work by turning backlog issues into verified pull requests using AI agents.

OpenWhispr is a cross‑platform desktop app that converts speech to text, provides live meeting transcription with speaker diarisation, and lets you talk to AI assistants (GPT‑5, Claude, Gemini, etc.) directly from your keyboard. It works offline with Whisper‑based models or via cloud services, stores notes locally (with optional cloud sync), and offers an API for programmatic use.

Graft is an open‑source TypeScript/Node.js tool that builds a local, markdown‑based knowledge graph of a codebase (summaries, key snippets, typed links) using tree‑sitter for structural parsing and optional LLM summarisation. The graph is cached in a `graft/` folder, refreshed only on file changes, and automatically wired into coding agents like Claude Code, Cursor, Codex, and Gemini. Benchmarks claim up to 46 % fewer tool calls, 42 % token savings, 60 % latency reduction, and a 12‑point correctness boost on SWE‑bench. Integration is via a single `graft init` command that writes agent‑specific instruction files; the CLI also offers search, visualization, and telemetry controls. The project is MIT‑licensed and targets faster, cheaper AI‑assisted development.

A library of 163 standardized scientific and research skills that enable AI agents to interact with scientific databases, Python packages, and laboratory tools.

An open-source native desktop client for DeepSeek Harness that integrates its Web UI and plugin system into a standalone application for Windows and macOS.

An AI-powered toolkit for automating patent disclosure writing, interpreting complex patents into plain language, and assisting in drafting responses to patent examination notices.

GenOffice is a cross‑platform, open‑source office suite (Docs, Sheets, Slides, PDF, Markdown) that opens and saves native Microsoft Office formats with byte‑preserving round‑trips. Each editor includes a built‑in AI panel for block‑level editing and tool‑calling agents. Users can sign in to Genspark for keyless model access or provide their own API keys for many major LLM providers. The suite is packaged as signed installers for macOS, Windows, and Linux, and is released under Apache‑2.0.

Codex CLI is a local coding agent from OpenAI that allows developers to to run AI-assisted coding directly from their terminal.

An adaptive web scraping framework that bypasses anti-bot systems and automatically relocates data elements when website designs change.

A set of rules and filters for AI coding agents to prevent the generation of generic "AI slop" in UI, copy, and code comments, ensuring output is lively and specific.

An open-source container and VM runtime for macOS written in Rust that provides a drop-in Docker replacement and isolated microVM sandboxes for AI agents.

A library of reusable research skills for AI agents to automate high-impact academic workflows, from paper reading and Nature-style polishing to peer review simulation and patent drafting.

A design guidance system for AI coding agents that uses a shared command vocabulary and 61 deterministic rules to prevent generic AI-generated frontend designs.

MCP for Blender is an open‑source bridge that lets LLMs (Claude, Codex, Cursor, etc.) control Blender via a socket‑based server (the MCP server) and a Blender add‑on. Install `uv`, configure the LLM client with a short JSON snippet, install the add‑on, and then you can ask the AI to create, edit, or inspect 3‑D scenes, run Python code, and pull assets from Poly Haven, Sketchfab, Poly Pizza, or AI‑generated model services. Safe‑mode can sandbox Python execution, and the server can run locally, via Docker, or with `pipx`. The project is packaged on PyPI and supported through a Discord community.

A high-performance serving framework for LLMs and multimodal models that provides low-latency, high-throughput inference across single GPUs and large distributed clusters.

An AI-powered workflow that converts documents into fully native, editable PowerPoint presentations with real shapes, charts, and tables.

A repository of markdown‑based AI‑agent personas (e.g., Frontend Developer, AI Engineer, UI Designer) with scripts to convert and install them into many AI‑coding assistants (Claude Code, Cursor, Gemini CLI, etc.). Provides ready‑made specialist prompts, workflows, and deliverable templates for rapid use across tools.

An open-source toolkit that implements a spec-driven development process for AI coding agents, ensuring software is built from executable specifications rather than ad-hoc prompts.

A smart AI router and proxy that saves 20-40% of tokens and provides automatic fallback between subscription, cheap, and free AI models for coding tools.

A Chrome extension that converts selected web content into editable office files like spreadsheets, slide decks, and HTML sites using user-provided AI models.

Caveman is a token‑saving wrapper for AI coding assistants. It provides a tiny “skill” that rewrites the assistant’s explanatory text in a terse style, and an optional local proxy that compresses large inputs (logs, JSON, diffs, etc.) before they reach the LLM provider. Benchmarks show up to 65 % fewer output tokens and ~33 % fewer input tokens, lowering cost and latency for agents like Claude Code, Codex, Gemini, and many others. Installation is a one‑liner for the skill and a global npm install for the proxy; the tool works with 30+ agents without altering their config files.

A high-performance CLI proxy that compresses bash output for AI agents, reducing input tokens by up to 90% through smart filtering and grouping.

An open-source SEO tool that integrates with AI agents via MCP, providing a pay-as-you-go alternative to traditional SEO suites for keyword research and site audits.

A secure bridge that lets the ChatGPT web app handle planning and review for Codex coding sessions, reducing API token consumption by using web subscription quotas.

An open-source, local-first design platform that enables coding agents to generate and render web prototypes, decks, and motion graphics based on a brand's design system.

A Python tool that converts PDFs, EPUBs, DOCX, Markdown, etc., into an Agent‑Skill bundle (SKILL.md + per‑chapter markdown files). The skill can be loaded by Copilot CLI, Amp, Claude Code, letting the LLM answer questions from the original book on‑demand while using 24‑51× fewer tokens than feeding the whole book as context.

TimesFM is Google Research’s decoder‑only foundation model for time‑series forecasting. Version 3.0 supports native multivariate forecasting with past and future covariates, zero‑shot performance, and quantile uncertainty. The library (`timesfm` on PyPI) provides PyTorch‑based inference, fine‑tuning examples (LoRA), and integration points for Google Cloud services. Code is Apache‑2.0; v3.0 weights are under a non‑commercial license.

Agent Reach is an open‑source Python CLI that equips any LLM‑based AI agent with ready‑made, auto‑updating ways to read webpages, fetch YouTube subtitles, query Reddit, pull GitHub data, and access many other platforms. It selects, installs, and health‑checks the best open‑source tools for each service, registers a `SKILL.md` for the agent, and provides a `doctor` command to diagnose channel status. The project is free, MIT‑licensed, and focuses on safety (local credential storage, no system changes without explicit `--system`).

A desktop app and browser extension that bridges ChatGPT to a user's local computer, enabling controlled access to files, shell commands, and Windows desktop automation via MCP.

A multi-agent orchestration configuration for Codex that uses GPT-6 Astra as the root manager and reviewer, and GPT-5.6 Luna for execution subagents.

A library of agent skills that enables AI to generate, inspect, and prepare CAD models and robot-description files for simulation and fabrication.

A software factory framework that turns AI coding agents into a virtual engineering team with specialized roles for product planning, architectural review, QA, and security audits.

A tool that removes over 20 common patterns of generic AI-generated writing to help authors maintain a personal, human voice in their AI-assisted drafts.

A collection of Reinforcement Learning robotic environments powered by the MuJoCo physics engine and the Gymnasium API for training AI agents in manipulation and navigation tasks.

An open-source proxy that lets you run popular coding agents using 50+ free, paid, or local LLM providers with automatic failover and token optimization.

Agent Skills is a library of 25 Markdown‑based engineering workflows (spec, plan, build, test, review, ship, etc.) that can be installed into many AI coding agents via a simple CLI. It provides slash commands, auto‑build mode, specialist personas, and reference checklists to make AI‑generated code follow senior‑engineer best practices across the entire development lifecycle.

A self-hosted developer control center for managing coding agents and automating engineering workflows across local, remote, and cloud backends.

A service and agent skill set to strip AI provenance marks and watermarks from text and various file formats for privacy and hygiene.

An autonomous AI penetration testing tool that uses multi-agent orchestration to find, validate with PoCs, and fix security vulnerabilities in applications.

A curated library of 216 numbered hand‑drawn illustration styles with ready‑made bilingual prompts, plus a simple web gallery. Users pick a style ID, add a subject, and get a prompt that works with text‑to‑image AIs, ensuring consistent visual tone without having to describe the style themselves.

An agent skill that uses a closed-loop cycle of image generation and AI critique to build visually impressive games, apps, or 3D scenes.

A collection of 11 AI skills for Claude Code and Codex that automate LinkedIn content creation, engagement, and humanization of AI-generated text.

A unified OpenAI-compatible API gateway that aggregates free tiers from dozens of LLM providers into a single endpoint with smart routing and failover.

lieflat‑gongwen is a data‑driven “skill” for Chinese AI agents that lets them generate official government‑style documents (work plans, research reports, party‑building material) by matching statistical style fingerprints extracted from a 1 M‑character corpus. It detects the document type, applies numeric style constraints, builds a skeleton, fills the text with an LLM, and runs a Python self‑check to ensure the output stays within the measured ranges. The repo includes the rule set, example drafts, and a non‑commercial license.

A cross-platform manager for AI coding agents and CLI tools that provides a unified interface to configure and configure and switch between tools like Claude Code, Codex, and Gemini CLI.

A local API traffic capture tool for AI coding agents that attributes token costs to specific request components like system prompts and tool schemas.

A high-capability coding agent with deep IDE, debugger, and OS integration that allows LLMs to perform complex software engineering tasks with high precision.