$ loading_
60320 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 web and local results privately to quickly find information and resources.
Triage security advisories, drafts, and GHSA reports with trusted evidence.
Diagnose OpenClaw node pairing, auth, routing, and connection issues.
Set up and use 1Password CLI for sign-in and secret injection.
Inspect, patch, validate, and publish OpenClaw GHSA advisories securely.
Audit and harden OpenClaw hosts for security and operational health.
Automatically add redacted agent transcripts to GitHub PRs or issues.
Run or recover OpenClaw macOS signing, notarization, and release promotion.
Manage Feishu document and file sharing, access permissions, and collaborators.
Capture frames or short clips from RTSP and ONVIF cameras.
Helps maintainers triage, clean up, and resolve OpenClaw secret scanning alerts.
Apply Django security best practices for safer auth, app defenses, and deployment.
Write, review, and refactor C++ with modern, safe, idiomatic standards.
Review Flutter and Dart code for architecture, performance, accessibility, and security.
Validate router and switch configs before deployment to catch security and network risks.
Audit production readiness locally and uncover risks before or after release.
Plan a reliable home or homelab network with practical setup guidance.
Learn production-ready Redis patterns for caching, concurrency, and messaging.
Troubleshoot BGP sessions, routing policy issues, and collect safe diagnostic evidence.
Set up and manage Pi-hole DNS, blocking, and home network resolution.
Review security risks for auth, inputs, secrets, APIs, and sensitive features.
Prevent destructive actions in production systems and autonomous agent workflows.
Audit repository assets, detect third-party code, and generate module-level verdict reports.
Run a pre-release verification loop for Quarkus builds, tests, scans, and reviews.
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。