发现 Skills

Collection of 130+ specialized Claude Code subagents for development tasks across languages, frameworks, infrastructure, and security

AI-assisted academic research workflows for literature review, paper writing, peer review, and research pipelines

End-to-end ETL pipeline for Harvard Art Museums API with SQL analytics and Streamlit visualization

Infrastructure-as-Code fundamentals for data engineering using Terraform to provision AWS resources (S3, EC2, IAM)

AI-driven autonomous penetration testing with Kali Linux tools, multi-phase attack planning, and human approval gates for high-risk actions

```markdown

```markdown

Detector and proof-of-concept LPE toolkit for CVE-2026-31431 ("Copy Fail"), a Linux kernel algif_aead page-cache scratch-write vulnerability enabling local privilege escalation.

A Claude Code skill that polishes Korean text written by AI so it reads like human-written — detects and surgically fixes 40+ sub-patterns of translationese, idiomatic quirks, and structural AI patterns.

```markdown

Transform vague ideas into implementation-ready specifications through structured interviewing. Use when user describes a new feature/product idea, has a problem to solve, needs to document requirements, or says things like "帮我梳理需求", "interview me about this", "细化这个想法", "spec out this project". Produces two artifacts - intent.md (technical spec for code agents) and overview.md (human-friendly summary for team).

Use when an agent needs to import OPML, parse RSS or Atom feeds, monitor new AI or technical articles, filter entries by keyword/date/author/category/language, score article quality, dedupe seen items, or evaluate RSS source quality for high-signal technology content discovery.

When the user wants to create competitor comparison or alternative pages for SEO and sales enablement. Also use when the user mentions 'alternative page,' 'vs page,' 'competitor comparison,' 'comparison page,' '[Product] vs [Product],' '[Product] alternative,' or 'competitive landing pages.' Covers four formats: singular alternative, plural alternatives, you vs competitor, and competitor vs competitor. Emphasizes deep research, modular content architecture, and varied section types beyond feature tables.

Use before installing dependencies, packages, or tools required for building project.

Tmux guide for running background tasks. Use this guide when: 1. Before any tasks that can potentially run for more than 2 minutes (e.g.: package installs, many tests, model training) 2. Before any tasks that are expected to run indefinitely in the background (e.g.: web servers, port forwarding) 3. When a bash tool reports `timeout after 120000ms` 4. When the user requests to run tasks in the background

Use the ClawdHub CLI to search, install, update, and publish agent skills from clawdhub.com. Use when you need to fetch new skills on the fly, sync installed skills to latest or a specific version, or publish new/updated skill folders with the npm-installed clawdhub CLI.

Use this skill when implementing security measures or conducting security audits. Provides OWASP Top 10 mitigations, authentication patterns, input validation strategies, and compliance guidelines. Ensures applications are secure against common vulnerabilities.

Complete GitHub expertise covering GitHub Actions, CI/CD workflows, automation, repository management, and best practices. Use when setting up GitHub Actions, creating workflows, managing pull requests, configuring automation (Dependabot, CodeQL), or implementing GitHub best practices. Includes workflow generators, templates, and production-ready configurations.

AGI-level autonomous controller — self-healing, self-optimizing, constitutionally governed by arifOS F1-F13

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

This skill MUST be invoked when an analysis of an extension's composition (CFE) is required: borrowed objects, interceptors, and to check whether snippets have been migrated. It SHOULD also be invoked before deploying the extension to production. Do NOT use this for extension validation — use cfe-validate.

Build a 1C external processing (EPF/ERF) from XML sources

Audit and improve CLAUDE.md files in repositories. Use when user asks to check, audit, update, improve, or fix CLAUDE.md files. Scans for all CLAUDE.md files, evaluates quality against templates, outputs quality report, then makes targeted updates. Also use when the user mentions "CLAUDE.md maintenance" or "project memory optimization".

Use this skill when a user asks to "create a hookify rule", "write a hook rule", "configure hookify", "add a hookify rule", or needs guidance on Hookify rule syntax and patterns.