An Excel processor that converts Korean addresses into English addresses and adds the result beside the original address column. It handles common Korean administrative areas, roads, building details, and delivery notes.
An HTML parser for Medium articles. It turns an article's page structure into JSON containing metadata, headings, paragraphs, images, code, quotes, lists, and other sections.
An English-to-Korean translator for structured article content represented as JSON. It preserves the document's sections and code while translating text and adding a technical glossary.
A formatter that turns translated article JSON into a readable Markdown document. It adds frontmatter, headings, body text, lists, code blocks, links, images, and inferred tags.
Translate English SRT subtitle files to Korean. Use when given an English SRT file path to translate subtitles into natural Korean and save to Downloads folder.
A WikiDocs page converter that saves the main content of a WikiDocs page as a Markdown file. WikiDocs is a website for publishing structured learning and reference documents.
A command that runs one or more project agents, either one after another or at the same time. Project agents are saved instructions for handling specific kinds of work.
A command that passes work through a chain of agents, where each agent receives the previous stage's result. The chain is written with arrows, such as one agent followed by another.
A command that compresses a long agent session by splitting it into work blocks and processing those blocks in parallel. It checks the merged result for missing information and stores the handoff separately for each session.
A command that compresses a long agent session into structured handoff files. A handoff records the work state so another session can continue, and replay checking looks for information lost during compression.
A weekly review command for improving an agent setup over time. It examines recent activity, repeated work, repeated mistakes, instruction conflicts, unused skills, and possible security issues.
Collect and synthesize opinions from multiple AI agents. Use when users say "summon the council", "ask other AIs", or want multiple AI perspectives on a question.
Read when you need a browser automation (QA, element interaction, screencapture/snapshot, network capture, so on), or have to work across user's logged-in accounts, apps (e.g. Slack, X, LinkedIn, etc.), memory, and browsing history.
Unified browser automation — the single entry point for driving a real web browser: opening/reading/ scraping pages, filling and submitting web forms, clicking through multi-step flows, taking screenshots or PDFs, downloading files from web pages, researching across many sites at once, and operating LOGGED-IN web…
A setup guide for running Claude Code inside GitHub Actions, GitHub's service for automated tasks connected to code repositories. It covers scheduled, manual, and issue-comment workflows, plus integrations with external tools.
Set up a web-based Claude Code conversation history viewer with FastAPI backend. This skill should be used when the user wants to view, search, or manage their Claude Code conversation history through a web interface, or when they ask about setting up a history viewer, conversation browser, or session manager for…
Call OpenAI Codex CLI (or ChatGPT CLI) from within Claude Code to leverage GPT models for cross-checking, code generation, or alternative perspectives. This skill should be used when the user requests to use Codex, ChatGPT, OpenAI, or GPT models from within Claude Code, or when cross-validation with another AI is…
INVOKE THIS SKILL when building ANY Deep Agents application. Covers createdeepagent(), harness architecture, SKILL.md format, and configuration options.
INVOKE THIS SKILL when your Deep Agent needs memory, persistence, or filesystem access. Covers StateBackend (ephemeral), StoreBackend (persistent), FilesystemMiddleware, and CompositeBackend for routing.
INVOKE THIS SKILL when using subagents, task planning, or human approval in Deep Agents. Covers SubAgentMiddleware, TodoList for planning, and HITL interrupts.
Extract text from images and PDFs using DeepSeek-OCR model with MLX. This skill should be used ONLY when the user explicitly mentions "딥식", "Deepseek", "DeepSeek", or similar variations (딥시크, 딥씨크, deepseek, etc.) in their OCR request. Do NOT use this skill for general OCR requests. Supports PNG, JPG, WEBP, BMP, GIF…
An audio-to-text add-on that uses ElevenLabs Scribe to turn recordings into written text, then corrects likely mistakes using your projects, notes, memory, and glossary. It can keep the original transcript and a record of corrections.
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.