Tag
The open-source AI Skill project geo-sleuth is available on GitHub. It can determine where a photo was taken, even without GPS data, road signs, or text, using clues such as ridgelines, bridge piers, shadows, and satellite maps. An official example achieved an error of just 2 meters, and it supports various AI agents, including Claude Code, Codex, and Cursor.
I've open-sourced a Claude Code Agent Skill that scrapes the complete comment section of a single Douyin video (including nested replies) by reusing real request templates and the page SDK's signing mechanism within a logged-in Chrome page. It requires no paid APIs or reverse engineering of a_bogus, does not trigger CAPTCHAs, and runs about 2.5 times as fast as browser automation solutions.
An open-source Agent Skill called match-cut packages the entire workflow for creating beat-synced, match-cut promotional videos for use with Claude Code and Codex. Simply provide product images and information to generate a product promotional video with one click.
开源了「神佬知识大赏」Agent Skill（Digital Human Koubo），只需一张照片加一句话即可全自动生成带口播、配音、字幕和包装的数字人短视频，已在 GitHub 上发布。
ste-zh 是一个开源 Agent Skill，借鉴 ASD-STE100 简化技术英语规范，约束 AI Agent 用清晰、无歧义的中文汇报工作结果，支持通过 skills CLI 或手动安装到 Claude Code 等环境。
开源 Agent Skill「android-adb-control」让 AI Agent 通过 ADB 远程操控安卓手机，支持观察界面、点击、滑动、输入等操作，并可配合 Tailscale 实现无线远程控制，兼容 Claude Code、Codex CLI、Pi Agent 等多个宿主。
开发者开源了桌面版 Git 可视化工具 Oil Git，支持查看文件改动、提交图、分支状态，并附带 CLI 与 Skill 供 Agent 读取仓库状态，已在 GitHub Pages 上线并提供 macOS/Windows 下载。
The learn-from-materials Agent Skill converts PDFs, EPUBs, Word docs, and PPTs into interactive learning websites with source tracing and knowledge base building.
A universal media downloader tool built as an AI agent skill for platforms like YouTube and Bilibili, with a GUI web version and support for personal deployment.
The author open-sources a 'lean-thinking' skill for AI coding agents that reduces token usage by up to 35% while maintaining quality, based on personal testing with Claude and Codex models.
An open-source Agent Skill and guide for transforming IP characters into minimalist, cute logos using AI image models, with specific design principles and a human-AI collaboration workflow.
Open-sourcing first-reader, a free agent skill that simulates human readers to provide feedback on drafts by analyzing passages and reporting reactions.
This post introduces a skill that enables users to share Blender 3D worlds on the Hugging Face Hub using an agent, with automatic contributions to a public dataset.
This article introduces Claude.md doctor, an open-source agent skill that examines and grades CLAUDE.md or AGENTS.md files, providing fixes and backtesting for session flaws.
An AI tool that writes marketing copy and strips out AI-written signs to produce human-like text, based on communication research.
The author open-sourced oil-motion, an Agent Skill for integrating continuous motion generated by AI video into web interactions such as scrolling and mouse, providing a complete solution covering generation, inspection, compression, and front-end integration.
An agent skill that forces LLMs to write documentation in ASD-STE100 Simplified Technical English, reducing STE violations by 72.9% across six Claude models.
An open-source agent skill classifies coding tasks into S/M/L/XL complexity classes to set appropriate execution depth, preventing overthinking simple fixes and enabling evidence-based reclassification.
A GitHub repository video-shotcraft provides 106 shot recipes, 162 styles, and 161 dynamic previews organized as Agent Skills for AI video generation using Remotion and compatible with Claude Code and Codex.
ClickHouse releases @clickhouse/rowbinary, a Node.js library for reading/writing RowBinary formats that also includes a skill file enabling coding agents to generate query-specific, high-performance parsers that run 1.5–3.4x faster than generic parsers.