Explore skills

A skill that optimizes and diagnoses Korean content for GEO (Generative Engine Optimization), improving its citation by generative AIs like ChatGPT and Gemini, based on the Princeton/IIT KDD 2024 GEO paper.

Knowledge base for AI assistants advising on OnlyFans talent agency operations. Covers chatter cost economics at 5-25 creator scale, P&L structure by operating model (chatter-only, assisted-AI, autonomous AI), revenue-leakage modeling, turnover-tax accounting, and the autonomous-vs-assisted threshold.

This skill provides quantitative ETF trend analysis, leveraging ETF fund flow data to determine stock market trends. It generates structured reports based on broad-based ETF flows, industry differentiation, cross-border ETF premiums, and 'national team' movements.

This skill allows users to create a virtual partner with specific personality and emotions. Users can provide photos, audio, and descriptions for Claude to act as a lively character, responding with injected emotions and personality.

Manually capture missed insights from recent conversation. Use when thinking-tree's auto-recording missed something valuable. Accepts a natural language hint about what to capture.

Distill fragments into structured thought files — classify into existing files or synthesize new ones. Scans fragment pool for related insights, proposes actions, and executes. Accepts optional topic hint.

Clean the thinking-tree fragment pool — filter by preferences, deduplicate, and remove outdated entries. Uses AskUserQuestion for option-based interaction so user can review without typing.

Analyze disk space and find cleanable files. Use when user wants to clean disk, free up space, find large files, analyze disk usage, or asks about what's taking up space. Triggers on requests like "clean my disk", "free up space on C:", "find large files", "what's using my disk space", "disk cleanup".

Creates Python projects with proper structure, virtual environments, and dependency management. Use when users request to create a new Python project, set up a Python development environment, or initialize a Python application with standard tooling.

Practical memory architecture for OpenClaw agents. Use when an agent needs to remember commitments across sessions, retrieve prior context, prevent repeat mistakes, stage durable-memory updates safely, compile reusable procedures, or improve memory hygiene over time. Supports a 3-tier adoption path: Core, Recommended, and Advanced.

The universal Jarvis core v3.1, adaptable to most platforms, is a qualitative upgrade from a tool-based assistant to a soulful partner. It integrates a complete emotional system and a built-in tiered recall system, enabling AI to go from understanding humans to being like humans.

This repository is a collection of skills I use for my open-source contributions.