Research & Data

Rank Java files, methods, and REST API endpoints by a deterministic Composite Hotspot Score (recency-weighted churn × cognitive complexity × JaCoCo coverage gap) to prioritize test generation. Also a Claude Code skill/plugin.

macOS disk cleanup + tiered cold-storage archive (NAS/external/iCloud) as a Claude Skill

Claude powered advisor system for option traders

Score any Claude Code skill against a research-backed rubric derived from the top 9 most-starred skill repos on GitHub

A Claude Code skill that turns an academic paper PDF into a single self-contained interactive HTML explainer (Brain Dead Mode + Normal Mode, with approval gates). Forked workflow from paraschopra/make-pages-interactive.

AI-assisted ecommerce operations workflow demo with simulated data, SKU risk scoring, tests, and a reusable Skill.

Portable QA skill pack for AI coding agents: FE/API test-case prep (AC/EC + Swagger coverage review), Playwright ticket testing, defect filing, and Jira retest. Works in Claude Code, Cursor, Codex, GitHub Copilot, Gemini CLI, Windsurf, Cline, and Agent Skills–compatible tools.

Open-source Codex skills for creating learning-aligned assessment questions and generating scaffolded learner feedback.

Bootstrap your own personal AI operating system -- a 4-tool stack (Claude Code + Lark Base + Obsidian + Lark Tasks) plus a character system that remembers you across sessions. Install: npx skills add breakthrough-edu/personal-ai-os-bootstrap

Claude Code skill inspired by DeepTutor: code-first deep research + Socratic tutor with a persistent learning workspace.

A Codex skill for World Cup match prediction and market signal analysis.

A Codex skill for organizing academic paper projects with folders, specialist threads, and handoff files.

Claude Agent Skill for operations research: LP/MIP formulation, network flows, scheduling, queueing, inventory, simulation, and Markov models — solved in Python (PuLP · Pyomo · SciPy · OR-Tools · NetworkX)

Agent Governance & Assurance — AI Agent Skill security scanner. Rule-first detection of malicious agent skills with optional LLM deep analysis. YAML-based plugin system, MalSkillBench integration.

12 AI agents in your git workflow. Catches secrets, injection flaws, N+1 queries, and bad migrations on every commit. Generates production-ready code, product wikis, and codebase maps on demand. Any language. Free.

A Claude Code skill that turns Everything (es.exe) into Claude's default file-search tool on Windows.

Personal LLM Wiki + Pattern Library + Routine v2.1 — Karpathy's pattern applied to agentic-AI ecosystem analysis (56-wiki corpus, 39 confirmed patterns, MIT)

Agent skill repository: databricks/databricks-agent-skills

Shared infrastructure for AI-assisted research workflows: Claude Code skills, conventions, methodology docs, and tools.

Claude Code skill: search Perplexity, Bing Copilot, and Google AI simultaneously — synthesized AI answers, not just links

Investment-research skills for Claude Code: top-down macro-aware framework, bilingual EN/CN NL triggers, Telegram alerts (2-min cron + 1-3s webhook). For personal-finance buy-side use (swing/position/LEAPS) — not HFT. See NEXT-STEPS.md for roadmap.

AI-assisted screening for data-integrity fingerprints in scientific papers · 学术论文数据完整性筛查流水线(耿同学)

自媒体多平台数据分析 → PPVI 浅色 HTML 报告页 · Claude Code skill

Discover and compare open-source Agent Skills, tools & MCP servers — with quality scoring, trending analysis, and automated GitHub sync