Research & Data

Runtime-agnostic agent skill for evidence-bound supplier discovery reports. Works with Claude Code, Cursor, OpenCode, Codex, Gemini CLI.

Agent Skill for bird-dogging — driving cross-team work items to completion across pluggable trackers (Jira, GitHub Issues, …) with local-LLM triage and customizable status reports.

🤖 Local-assist BOSS Zhipin CLI for AI agents — search, welfare filtering, shortlist, JSON-envelope output; low-risk & compliant by default.

A curated hub of autonomous-research skills & agents — from idea to paper, on autopilot. | 自主科研技能与智能体精选库 —— 从灵感到论文，全程自动驾驶。

PM knowledge base skill for Claude Code. Karpathy's LLM Wiki pattern with supersession, privacy, crystallize, and qmd hybrid search.

A workflow and safety harness: plan, implement, review, and guard AI coding sessions.

AI agent skill for TestRail REST API - zero dependencies, security-friendly alternative to MCP servers

한국 담배 규제를 지키며 맛 특징을 살리고 중복을 피하는 담배 제품명 작명 Claude Agent Skill

Codex skill: closed-loop research workflow with multi-agent adversarial gap discovery, Semantic Scholar validation, and 7-phase research cycle

Claude Code Agent Skill for unified web search across multiple providers

to collect useful codex skills for academic research

Official COROS MCP server and Agent skills for integrating sports and health data into AI clients.

Codex skill for Feishu daily work reports

Codex skill for AI Huokebao customer usage diagnosis, customer success SOPs, and business reports.

Convert PDFs to lean Markdown so Claude reads the cheap version — with a measured token and cost savings report. A Claude skill plus CLI.

Claude Skills Dashboard — all installed skills, searchable and filterable

Claude Code WeChat Integration - Sync conversations to WeChat via Stop hook

All-in-one Codex skills for research, planning, native execution, verification, and evidence

Search, read, and download curated agent skills via MCP for fast discovery and local use in agent workflows

Optimize Agent Skills with session data and static analysis to find trigger gaps, weak workflows, and P0/P1/P2 fixes

Portable DIY agent skills + MCP servers (MinerU PDF parsing, SciVerse literature search, learning-mechanics) with a one-command idempotent installer for Claude Code / Codex / Cursor / OpenCode.

A reusable agent skill for auditing listed-company financial reports, scoring financial quality, and producing red-flag analysis reports.

Analyze Claude Code source leak to inspect the CLI source map and study the exposed readable code from the npm package

Analyze BaZi charts with Claude Code, gather birth data, and generate four pillars, luck cycles, and practical命理 insights from classic texts