$ loading_
36505 Skills & MCP tools — copy a prompt, zero pre-install, into Claude Code / Codex / Cursor.
🎓 Skill = teach the AI how to do a task, copy & use; 🔌 MCP = connect the AI to an external tool/data source, usually needs a key.
Fetch web pages and convert them to Markdown for AI reading and analysis.
Connect to Notion to query databases, create pages, and manage workspace content in bulk.
Transcribe audio to text locally with Whisper CLI, no API key required.
Create, search, and manage Bear notes on macOS via grizzly CLI.
Monitor blogs and RSS/Atom feeds to catch new updates quickly.
Read, edit, and write Feishu documents for faster cloud collaboration.
Create and review technical docs and agent instruction files in repositories.
Search the web, extract content, and organize research efficiently.
Convert text to speech locally and offline with sherpa-onnx, no cloud needed.
Convert text into high-quality speech with a macOS say-like workflow.
Draft OpenClaw release announcements and testing guidance from release evidence.
Create, organize, search, and export Apple Notes from the macOS command line.
Maintain an Obsidian-friendly knowledge vault with links, metadata, and structure.
Find meme templates, suggest formats, and generate shareable meme images.
Summarize or transcribe URLs, videos, PDFs, and local files quickly.
Regenerate OpenClaw release changelog sections from Git history before releases.
Manage Obsidian vault notes, tasks, links, and plugins via CLI.
Maintain an OpenClaw memory wiki with deterministic, source-backed page updates.
Transcribe audio into text or speaker-separated transcripts with OpenAI APIs.
Refactor OpenClaw docs pages with source-checked preservation, clearer structure, and verification.
Create, revise, and validate AI skills before deployment.
Conduct multi-source web research and produce cited, source-attributed reports.
Forge complete OpenClaw lobster personas with lore, rules, names, and avatar prompts.
Conduct systematic literature reviews with search planning, screening, synthesis, and citation checks.
从访谈内容留存、转写回看，到主题归纳与产品建议，一套串起用户研究整理流程。
从流量采集、营销归因到数仓查询与看板输出，一套串起自动取数和分析闭环。
从读取设计稿、提取上下文到映射组件库与同步评审，这套组合很适合设计到前端落地的协作链路。
In-depth guides on Skill / MCP setup and security — beyond the directory listings
普通聊天 AI 很难真正读懂大型仓库，因为它缺少结构、上下文和可验证的检索路径。本文从 0 到 1 搭一个只面向本地代码库的 Cursor 助手：能查结构、追调用链、看 diff、给重构建议，还能尽量压住幻觉。
Agent-hop 是一个基于 Rust 的 TUI，支持在进行中的 Claude Code 聊天里切换到其他编程代理，尽量不丢失上下文。项目作者提到，它最早源自一个简历与搜索工具，后来因为使用过程中的阻碍，单独做成了现在的界面。
Harden.run 表示，针对能够生成任意代码的 Coding Agent，单靠传统红队测试并不够。他们通过后训练网络安全小模型、调整推理方式，并结合 inline reference monitoring 等程序分析手段，在 LinuxArena 和 SleightBench 上取得了超过 GPT5.5-xhigh 的成绩。