$ loading_
35650 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.
Search the live web and codebase in real time for company research and information retrieval.
Connect to and manage Supabase projects, perform table operations, and analyze queries.
Connect to Stripe APIs to handle payments, orders, and finance-related operations.
Help development and operations teams investigate errors, manage issues, and analyze performance monitoring data.
Search and analyze past session logs with jq for context and insights.
Search and analyze GitHub archive issues, PRs, and repo workflow status.
Search places, fetch details, reviews, and structured JSON from Google Places.
Search the web, extract content, and organize research efficiently.
Search and analyze Slack archives, count threads, and support repo workflows.
Generate spectrograms and feature-panel visuals from audio for analysis and presentation.
Search, export, sync, and analyze archived Notion content and related repo data.
Search and manage Granola archives, sync status, notes, transcripts, and counts.
Transcribe audio into text or speaker-separated transcripts with OpenAI APIs.
Summarize CodexBar local cost logs by model for Codex or Claude.
Optimize energy procurement, tariffs, PPAs, and multi-site energy cost strategies.
Optimize production schedules, sequencing, bottlenecks, changeovers, and disruption response.
Post, search, read timelines, and analyze data via the X API.
Verify revenue, pricing, refunds, and billing truth with evidence-backed answers.
Build practical monitoring dashboards that answer real operator questions.
Research current facts, compare options, and produce evidence-based recommendations.
Create consistent fundraising documents, financial projections, and investor-facing materials.
Search official USPTO records and organize reproducible patent and trademark research.
Apply PyTorch best practices for robust, efficient, reproducible deep learning pipelines.
Audit, plan, and implement SEO improvements for better search visibility.
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。