AI for Beginners Curriculum: Học Trí tuệ Nhân tạo

prompts.chat is the largest open-source prompt library for AI, formerly called Awesome ChatGPT Prompts. It hosts curated prompts in CSV and Markdown, available as a public website, Hugging Face dataset, or self-hosted instance. The project supports multiple LLM providers including ChatGPT, Claude, Gemini, Llama, and Mistral. Self-hosting uses a Next.js setup wizard that configures authentication via GitHub, Google, or Azure AD, with PostgreSQL as the recommended database. CLI access, an MCP server, and a Claude Code plugin extend its reach into developer workflows. The codebase is MIT-licensed while prompt data falls under CC0. Its 166k GitHub stars make it an AI resource on the platform with 166k GitHub stars, and it has been cited by Harvard, Columbia, and Forbes.

Supervision is a Python toolkit developed by Roboflow that offers reusable components for common computer vision workflows, aiming to simplify tasks from data loading to real-time analysis. It provides model-agnostic connectors for integrating various classification, detection, and segmentation models from libraries like Ultralytics, Transformers, or MMDetection. The library also includes customizable annotators for visualizing model outputs and a set of utilities for managing datasets, enabling operations like loading, splitting, merging, and saving data in formats such as COCO, YOLO, and Pascal VOC. With 48888 stars and 4617 forks, Supervision is designed for Python 3.10 environments and above.

screenpipe is a source-available application that continuously records a user's screen and audio, creating a local, searchable memory of their computer activity. It captures screen changes via an event-driven mechanism using the OS accessibility tree or OCR, transcribes audio locally with Whisper or via Deepgram, and stores all data in a local SQLite database. The tool functions as an MCP server, allowing AI assistants like Claude Desktop or Cursor to query screen history, and offers a "Pipes" plugin system for scheduled AI agents defined in markdown files to automate tasks based on captured data. It emphasizes privacy with 100% local data storage by default, supports local AI models, and provides deterministic, OS-level data permissions for AI agents via YAML frontmatter. The desktop app is available for macOS, Windows, and Linux, with a subscription model for signed versions, while the source is available for personal, non-commercial use.

AutoGPT is an open-source platform for building, deploying, and running AI agents that automate complete workflows. Users can define tasks in plain English using AutoPilot or visually construct steps with the Build interface. The project offers both a managed, hosted platform that handles infrastructure and model access, and a self-hosting option where users provide their own infrastructure and API keys. AutoGPT agents are designed to automate tasks across various domains, including executive operations, sales, and engineering, and integrate with over 45 platforms like Gmail and GitHub. While its `autogpt_platform/` component is licensed under Polyform Shield, the original `classic/` agent and other parts of the project use the MIT License.