Explore the directory

An all-in-one Python framework for semantic search, LLM orchestration and language-model workflows, built around an embeddings database that combines vector search with graph and relational storage.

• RAG
• Databases

An open-source, all-in-one backend platform for agentic coding that gives a coding agent a database, authentication, storage, compute, hosting and an AI gateway to ship full-stack apps end to end.

• Backend
• Hosting

A two-file embeddable C/C++ network library providing a full stack of HTTP, WebSocket, MQTT and TLS for microcontrollers and desktop operating systems alike.

• Libraries

An open-source library of 98 skills across 23 categories that equip AI coding agents to autonomously conduct AI research end to end, from literature survey and ideation through experiments to paper writing.

• Agent skills

A lightweight, open-source gateway that routes requests to more than 1,600 language, vision, audio and image models through one API, with automatic retries, load balancing, guardrails and an MCP gateway.

• LLM gateways

An idiomatic, open-source Java library for building LLM-powered applications on the JVM, with a unified API over more than 20 model providers and more than 30 vector stores.

• Agent frameworks

A Go framework for building LLM applications and agents, drawing on ideas from LangChain and Google's Agent Development Kit while following Go conventions, with a graph-based composition layer.

• Agent frameworks

An open-source, AI-powered code-review agent that runs as a GitHub Action, a local CLI or a webhook, reviewing, improving and answering questions about a pull request with a single LLM call per tool.

• Git
• Code quality

A Python library and local-first memory runtime that gives coding assistants and other agents one portable memory layer across apps and devices, stored as readable, Git-versioned Markdown.

• Agent memory

A history-rewriting tool for Git, recommended by the Git project itself in place of git filter-branch, that also works as a library for building custom repository-rewriting tools.

• Git

The open-source components of NVIDIA's TensorRT SDK for high-performance deep-learning inference, including plugins, an ONNX parser, sample applications and import-workflow guides.

• Inference

An open-source design system built on React and StyleX that ships accessible components, theming, and a CLI meant to be used the same way by people and AI coding assistants.

• UI kits

The official TypeScript SDK for building Model Context Protocol servers and clients, running on Node.js, Bun and Deno, with optional middleware adapters for common web frameworks.

• MCP servers

Developer-first, graph-native infrastructure that turns fragmented enterprise data into a queryable context graph and knowledge graph, with deterministic reasoning and end-to-end decision provenance.

• Agent memory
• Data engineering

A curated collection of more than 10,000 creative text prompts for Google's Nano Banana Pro image-generation model, organized by use case, visual style and subject, in 16 languages.

• Generative media
• Curated lists

A cross-platform 2D game engine for the Go programming language, handling graphics, input and audio so games can be developed once and deployed to desktop, mobile, web and consoles.

• Libraries

A terminal-native AI coding assistant that reads and writes code, runs commands, manages Git and keeps a persistent memory of a project across sessions while connecting to any mainstream LLM provider.

• Coding agents
• Terminals

A library of more than 20 large language models implemented from scratch in plain PyTorch, with recipes to pretrain, fine-tune and deploy them at any scale from one GPU to a thousand.

• Models

A Claude skill that writes precise, optimized prompts for other AI tools by detecting the target tool, extracting nine dimensions of intent, and asking a few clarifying questions before applying a matching prompt structure.

• Agent skills

An open-source, Chromium-based second browser built specifically for AI agents, importing a user's real logins so Claude Code, Codex and other agents can run web tasks in parallel tabs while sessions record for replay.

• Browser automation

An open, multi-language framework for building and orchestrating production-grade AI agents and multi-agent workflows in Python, .NET and Go, with graph-based orchestration patterns and Foundry-hosted deployment.

• Agent frameworks

A desktop and mobile web interface for using Claude Code, Cursor CLI and Codex remotely, with a chat interface, an integrated shell, a file explorer, a Git explorer and browser sessions for agent-driven tasks.

• Coding agents

An open-source Chrome and Edge extension for AI-powered web automation, running multi-agent workflows entirely in the browser with a user's own LLM API keys as a free alternative to OpenAI Operator.

• Browser automation

An application performance management tool that traces transactions across large, distributed Java, PHP and Python systems to map service topology and pinpoint bottlenecks.

• Monitoring