Directory

Explore the directory

Tools, libraries, services and references, each reviewed before it goes public. 6,821 resources listed.

6,821 listings

  • A routing layer for building Mixture-of-Models systems across heterogeneous LLM infrastructure. It evaluates request signals, user preferences, and application policies to select or compose the right model path for each request.

  • A curated collection of Agent Skills reflecting one maintainer's preferences and best practices, aimed at developers working mainly in Vite, Nuxt, and Vue.

  • SWE-bench

    swebench.com

    A benchmark that evaluates whether a language model can generate a patch resolving a real GitHub issue, given the codebase and the issue description.

  • Magic MCP

    21st.dev

    A compatibility proxy for the 21st MCP server, which searches over 10,000 React and Tailwind components, generates new UI with AI, and lets a user publish their own.

  • MCP Atlassian

    mcp-atlassian.soomiles.com

    An MCP server that gives an AI assistant tools to search, read, and update Atlassian Confluence and Jira, on both Cloud and Server/Data Center deployments.

  • x-transformers

    github.com

    A concise but fully-featured PyTorch transformer implementation bundling many experimental architecture variants and training techniques from published papers.

  • Cowart

    github.com

    A native infinite-canvas widget plugin for Codex, built on tldraw, for sketching ideas, annotating images, and generating and iterating images, HTML, and slide decks with Codex.

  • DevOpsGPT

    kuafuai.net

    A multi-agent system that combines a large language model with DevOps tooling to convert natural-language requirements into working, deployed software.

  • A curated collection of coding, system design, and machine learning interview questions from FAANG, frontier AI labs, and AI-first startups, updated for 2025-2026.

  • TrueForge

    trueforge.dev

    An open-source agent harness that runs the full agent execution loop, including model calls, MCP tools, skills, sandboxing, and approvals, behind a chat UI or an API.

  • Kaku

    kaku.fun

    A macOS terminal built for AI coding, based on WezTerm, that ships with fonts, themes, shell integration, and Mac shortcuts already configured.

  • keyd

    github.com

    A system-wide Linux key remapping daemon that uses kernel input primitives, with per-application layer overrides and a v2 configuration format.

  • KServe

    kserve.github.io

    A standardized, distributed inference platform for deploying both generative and predictive AI models at scale on Kubernetes, incubating under the CNCF.

  • Enchanted

    github.com

    An open-source, Ollama-compatible macOS, iOS, and visionOS app for chatting with privately hosted language models such as Llama 2, Mistral, or Vicuna.

  • agents-cli

    google.github.io

    A CLI and set of skills that teach a coding assistant to build, evaluate, and deploy AI agents on Google's Gemini Enterprise Agent Platform.

  • VLM-R1

    github.com

    A research project applying R1-style reinforcement learning to a vision-language model, reporting better out-of-domain generalization than supervised fine-tuning.

  • OpenCursor

    github.com

    An open-source, Cursor-like AI coding agent extension for VS Code that connects to multiple LLM providers, including local models, for chat-based code editing.

  • BrowserAct

    browseract.com

    A browser automation tool built for AI agents that gets past anti-bot defenses, hands off to a human when stuck, and isolates parallel tasks and accounts.

  • A large, continuously updated guide connecting Claude Code's documented behavior to engineering decisions, covering agent harness design, security, and team adoption.

  • LoopX

    loopx-project.github.io

    An open, local-first control plane that keeps long-horizon coding and research agents working across sessions, tracking goals, decisions, and evidence.

  • LobsterAI

    lobsterai.youdao.com

    A desktop-grade AI agent from NetEase Youdao that operates a real desktop, taking tasks from local chat or from WeChat, Feishu, DingTalk, and Telegram.

  • Cactus

    github.com

    A hybrid edge-cloud inference engine for running language, vision-language, and speech models directly on mobile devices and wearables.

  • Fontsource

    fontsource.org

    A monorepo that bundles self-hostable open-source fonts, mostly mirrored from Google Fonts, into individual, version-locked NPM packages.

  • ZCF

    zcf.ufomiao.com

    A zero-config, one-click setup CLI for Claude Code and Codex that installs workflows, an intelligent agent system, and API or CCR and MCP configuration.