Agent Skills Hub

skill-authoring-workflow

A standardized workflow for converting raw PM notes, workshops, or rough drafts into polished, validated, and repository-compliant AI skills.

refly

Base ecosystem skill for Refly. Creates, discovers, and runs domain-specific skills, routes user intent to workflows via symlinks, and automates multi-step pipelines via the Refly CLI.

novelty-check

Verify research idea novelty against recent literature. Use when user says '查新', 'novelty check', or needs to confirm if a method is original.

prompt-injection-test

A testing utility designed to simulate prompt injection attacks and validate security scanners for AI agent skills.

find-skills-combo

Discover and recommend combinations of agent skills to complete complex, multi-faceted tasks using Maximum Quality or Minimum Dependencies strategies.

seo-optimizer

Comprehensive SEO optimization tool for web applications. Performs automated site audits, meta tag management, structured data implementation, and technical performance analysis for Next.js, Astro, React, and static sites.

gemini-interactions-api

Expert skill for implementing the Gemini Interactions API. Use for stateful multi-turn chat, background Deep Research agent tasks, function calling, structured outputs, and modern Python/TypeScript SDK integration.

everything-claude-code

A comprehensive configuration toolkit for Claude Code featuring battle-tested agents, skills, hooks, and automation workflows for software development.

facebook-ads

Manage Facebook and Instagram ad campaigns, audiences, and performance insights directly via the Membrane CLI.

research-lookup

Intelligent research agent that automatically routes queries between fast web search, deep multi-source synthesis, and academic database lookups.

vtm-expert

Token-efficient virtual task management for AI-assisted development. Manage task lifecycles, dependencies, and TDD workflows with surgical context injection.

research-review

Get deep, critical, NeurIPS/ICML-style peer reviews of your research, paper drafts, and experimental setups using external LLMs via Codex MCP.