A skills set and methodology for coding agents: clarify the task, plan, write tests, then code. The agent works with more discipline.
The AI movement's repository database
The best AI repositories
224 living projects — from agent frameworks to prompt-injection defense. Each comes with our own note: why it matters, who it is for and how to start in three steps.
Stars, forks and activity refresh daily via the GitHub API · updated October 6, 2026
- repositories
- 224
- categories
- 11
- stars in total
- 11M
- new stars in 30 days
- +235K
Last 30 days
Growing fastest
The most new stars over the last 30 days.
A service that turns any website into clean Markdown or structured data for models. It can search, scrape, crawl a whole site and even click through a page.
A small Microsoft utility that converts PDF, Word, PowerPoint, Excel and other files into Markdown that is easy to hand to a language model.
An open coding agent for the terminal with a desktop app. It is not tied to one provider: plug in whichever models you have.
A toolkit for spec-driven development: principles and a spec first, then a plan and tasks, and only then code. Works with several coding agents.
OpenAI's lightweight coding agent that runs in the terminal and works on your project's code. Written in Rust.
Categories
Agent frameworksFrameworks and orchestration: what agents and multi-agent systems are built from.23 repositoriesCoding agents & CLIAgents that write code in the terminal and editor, and the methods for working with them.21 repositoriesMCPThe Model Context Protocol spec, official SDKs and the best servers.20 repositoriesOpen models & inferenceOpen weights, local runs, fast inference and fine-tuning.38 repositoriesRAG & vector DBsSearch over your own data: document parsing, crawlers, vector databases, knowledge graphs.19 repositoriesEvals & observabilityHow to measure a model or agent and see what happens inside it.17 repositoriesPrompting & contextPrompts, structured output, skills and context engineering.16 repositoriesVoice, video, multimodalSpeech recognition and synthesis, voice agents, images and video.19 repositoriesAutomationNo-code workflows, browser and computer-use agents.15 repositoriesAwesome lists & learningCourses, books and awesome lists worth starting with.21 repositoriesSecurityPrompt injection, guardrails, red teaming and agent scanners.15 repositoriesMissing your favorite?Suggest a repository — we will review it and add it if it earns a spot.Suggest a repo →
Catalog
43 repositories
A service that turns any website into clean Markdown or structured data for models. It can search, scrape, crawl a whole site and even click through a page.
Anthropic's official collection of Agent Skills: folders with instructions and scripts that Claude loads for a task. A good reference for writing your own.
Formerly Awesome ChatGPT Prompts: a huge community library of prompts plus an interactive book on prompting. You can deploy your own private copy.
The core library for working with models: one codebase for text, vision, audio and multimodal tasks, for both inference and training. Most new open models ship through it.
The best-known toolkit for LLM apps: one interface to dozens of models, plus tools, chains and agents. You can swap providers without rewriting the app.
A library that lets an agent drive a browser: open sites, click, fill in forms. Works with a local browser or a cloud one.
An animation engine Grant Sanderson uses to make 3Blue1Brown videos, including the well-known neural network ones. It is not an AI library but a tool for explaining math and ML visually.
A tool for running and fine-tuning models while saving memory: there is a desktop app, a web UI and a library. Fine-tuning works on a modest GPU.
An IBM library for document parsing: PDF, Office, HTML and images become one unified structure from which Markdown is easy to get. Supports vision models for hard pages.
More than 60 PyTorch implementations of research papers with notes next to the code: transformers, optimizers, GANs, reinforcement learning. The website shows code and notes side by side.
A memory layer for agents: it remembers facts about the user across sessions and pulls up the right ones at the right moment, keeping context small.
A high-level deep learning library that runs on top of JAX, TensorFlow or PyTorch: write a model once and pick the backend that suits the job.
Feeds fresh, version-specific library docs into the prompt so the model stops inventing outdated APIs. Works as an MCP server or through a CLI with a skill.
A low-level library for agents modelled as a state graph. It gives you long-running flows, saved progress and pauses for human approval.
Meta's classic library for fast similarity search and clustering of vectors. It is not a database but an index engine that many other systems are built on.
Google's library for extracting structured data from text with LLMs: each field is tied to a spot in the source, and results can be viewed in an interactive visualization.
The standard PyTorch library for diffusion models: ready-made pipelines for images, video and audio that run in a few lines.
A tiny Hugging Face library where the agent solves a task by writing its steps as Python code instead of JSON tool calls. Works with many models.
A layer on top of PyTorch that reduces training a network to a few lines. It comes with the well-known free fast.ai course, which makes it a handy place to start.
The AI SDK from the makers of Next.js: one TypeScript interface to models, streaming, tool calling and React hooks. Switching providers takes one line.
Whisper on the CTranslate2 engine: same output, noticeably faster and lighter on memory, including 8-bit mode.
A fast, memory-efficient implementation of attention that large-model training and inference on GPUs lean on. The result is exact, with no approximation.
A Postgres extension that adds a vector type and nearest-neighbor search. It lets you keep embeddings next to regular data without running a separate database.
A library for working with datasets: one line loads thousands of ready-made sets from the Hub or your local files and processes them fast. It can stream data without a full download.
A language for controlling generation: you mix text and constraints right in Python, and the model must produce exactly the required format, including schema-bound JSON.
A parameter-efficient fine-tuning library: LoRA and related methods train only a small fraction of the weights. Large models fine-tune on consumer hardware.
Hugging Face's library for post-training models: SFT, DPO, GRPO and reward-model training. The base for shaping model behavior after pretraining.
A tiny autograd engine with a small neural net on top and a PyTorch-like API. Very little code, yet it shows exactly how backpropagation works.
A unified interface to different generative AI providers: you only change the model string, such as `openai:gpt-4o` to `anthropic:...`, and the rest of the code stays the same.
A metrics library for evaluating RAG and other LLM apps: it checks how well an answer relies on retrieved sources. Can also generate test sets.
Did we miss something?
Suggest a repository
Send a GitHub link and a few words on why it belongs here. Every suggestion is reviewed by hand — not everything gets in.