File Ops
File manipulation and management
smart-commit
by toyi
Analyzes all uncommitted changes, groups them by theme (directory, semantic content, timestamps), and creates multiple focused commits. Use /smart-commit to create all commits automatically, or /smart-commit step to approve each commit one by one.
sync
by biruk741
Synchronize AI documentation across Claude Code, Cursor, and Codex. Detects drift between equivalent files and brings them into alignment while respecting each tool's format requirements. Use after updating docs for one agent to propagate changes to others.
ghrepo
by JsonLee12138
CLI tool for GitHub repository contents - browse, inspect, download, create, update, and delete files in any GitHub repository without cloning. Use when the agent needs to: (1) List directory contents of a remote GitHub repo, (2) Read/cat file contents from GitHub, (3) Download files or directories from GitHub to local filesystem, (4) Check file/directory metadata (type, SHA, size), (5) Create or update files in a GitHub repo (with commit message), (6) Delete files from a GitHub repo (with commit message), (7) Verify GitHub token authentication, (8) Initialize a project with AGENTS.md for a target repo. Triggers: "github repo", "remote file", "ghrepo", "browse repo", "download from github", "upload to github", "delete from github", "repo contents", "init repo".
deployment
by alexwelcing
Deploy, debug, and maintain services on Render and containerized platforms
celeborn
by Git-Scotch
Long-term context substrate for coding agents. Use to give a repository a small, persistent on-disk memory so sessions stay light, survive compaction, and resume cheaply across days. Trigger when the user says "set up long-term memory", "I keep losing context", "remember this across sessions", "make this project resumable", "rehydrate", "checkpoint the work", "hand this off", or when starting/resuming work in a repo that has a .context/ directory. Also use when context is getting heavy and you should offload state to disk.
taptapmaker-game-data-sync
by boomber3
将开发者在 UrhoX 游戏内编辑器中创作的数据同步到项目代码中。适用于以下场景： (1) 用户说"为我同步关卡数据"、"配置关卡 JSON"、"更新关卡"、"导入我的关卡" (2) 用户提供了 @Levelconfig.json 或粘贴了关卡 JSON 文本 (3) 用户反馈关卡在游戏中显示为空白 (4) 用户需要将游戏内编辑的角色属性、道具数据库、波次配置、对话脚本等数据同步到项目 核心工作：把 JSON 数据转换为 Lua 嵌入模块（scripts/XxxData.lua）， 并确保游戏通过 require("XxxData") 加载，而非 File:ReadString()。
tech-assessment
by tmaffia
Generate tech assessment reports for mentor/mentee relationships tracking growth, goals, and action items over time. Use when the mentor says "tech-assessment", "/tech-assessment", or "tech assessment" and provides meeting notes, a transcript, or a summary. The skill produces an updated overview.md and a dated assessment.md archival file.
ssh-remote
by leobrival
SSH Remote MCP expert for managing remote Ubuntu servers. Use when users need to connect via SSH, execute remote commands, transfer files via SFTP, or manage multiple server connections.
docx
by CodingHeader
"Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks"
architecture
by vilin1927
Architectural decision-making framework combining requirements analysis, trade-off evaluation, ADR documentation, Clean Architecture, DDD principles, and code quality guidelines. Use when making architecture decisions, analyzing system design, writing code, or evaluating trade-offs.
coco-cli
by yongsun-park
PCB design file analysis CLI (ODB++, Cadence .brd/.sip). Query layers/nets/components, measure feature distances, analyze net clearances, render PNG images.
File Organizer
by yousufjoyian
Claude agent skills and workspace configuration
Digital Immortality — 數位永生 Skill (v2.0)
by l12203685
Three-layer loop for any automated system: L1 Execute (do work) → L2 Evaluate (audit quality + coverage) → L3 Evolve (modify own execution rules). Execute without Evaluate+Evolve = dead loop. Same structure as Edward's belief update: expose → review → extract rule → write to system.
search-analyze-report-template
by djankies
[REPLACE] Search for patterns, analyze findings, generate structured reports. Use when [REPLACE with specific triggers].
dockerfile-generator
by karchtho
Generate production-ready Dockerfiles with best practices for various tech stacks. Covers multi-stage builds, optimization, security, and language-specific patterns (Node.js, Python, Go, Rust, Java). Use when creating new application containers.
vitepress-migration
by huguangju
Migrate existing Markdown documents to VitePress with proper routing, navigation, and Mermaid diagram support. Use when integrating legacy Markdown (.md) files into a VitePress project, converting documentation from Obsidian/Notion/other platforms, or setting up structured documentation sites. Triggers on requests involving Markdown migration, VitePress setup, documentation restructuring, or Mermaid diagram rendering in VitePress.
ttbrowserkit
by pyzhangxiang
浏览器自动化工具包。用于小红书等网站的稳定内容抓取、搜索、登录。采用 per-operation Chrome 生命周期，不会卡住。当需要浏览器自动化抓取小红书内容时触发。
brave-search
by blackhaj
Web search and content extraction via Brave Search API. Use for searching documentation, facts, or any web content. Lightweight, no browser required.
canvas-design
by Jackiexiao
（中文）Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
box
by CodeDraig
Enterprise cloud content management and collaboration.
spark-context-curator
by baggiponte
"Ultra-fast, read-only codebase exploration and context curation for GPT-5.3 Codex Spark. Use when you need deep repository understanding without modifying anything: architecture mapping, flow tracing, ownership discovery, incident/code-review prep, or implementation planning. Triggers on requests like \"explore this codebase\", \"curate context\", \"map where X happens\", \"investigate before editing\", or \"read-only deep dive\"."
create-skill
by Cygnusfear
This skill should be used when users want to create a new skill (or update an existing skill) that extends your capabilities with specialized knowledge, workflows, or tool integrations.
antenna
by ClawReefAntenna
Inter-host OpenClaw session messaging over reachable HTTPS using built-in gateway webhook hooks. Use when: (1) sending a message from this OpenClaw instance to another host's OpenClaw session, (2) checking status/health of a remote OpenClaw peer, (3) managing the peer registry (adding/removing/listing known peers), (4) exchanging bootstrap trust material for new peers, (5) any cross-host agent communication that should NOT go through visible chat channels like Telegram/WhatsApp/Discord. Triggers: "send to <peer>", "message the other host", "antenna send", "antenna status", "antenna peers exchange", "cross-host message", "inter-host relay", "ping <peer>", "peer list".
offload2
by 4ier
Cognitive pressure release valve for agent execution. Use when internal reasoning state cannot be carried forward productively - conflicting assumptions, high reasoning pressure, or risk of compounding errors. Invoke to explicitly let go of unproductive internal state. This is a one-way, non-blocking, deliberately unreliable sink. The agent must behave identically whether offload succeeds or fails.