Prompts
Prompt engineering and templates
vox-best-practice
by fleek-fitness
"vox.ai development best practices. (1) Voice agent system prompt design, authoring, and refactoring (template, {{...}} variable injection, filler options, character normalization, tools/silence actions, testing/production). (2) vox MCP server integration for ChatGPT, Claude Desktop, Claude Code, Cursor, OpenCode, Codex, VS Code Copilot."
code-review:config-manager
by ridewind
This skill should be used when the user asks to "manage review skills config", "update review skills", "discover review skills", "manage review presets", "validate review config", or mentions managing code review skills configuration. Use for configuration management tasks.
dspy-ruby
by 8b-is
This skill should be used when working with DSPy.rb, a Ruby framework for building type-safe, composable LLM applications. Use this when implementing predictable AI features, creating LLM signatures and modules, configuring language model providers (OpenAI, Anthropic, Gemini, Ollama), building agent systems with tools, optimizing prompts, or testing LLM-powered functionality in Ruby applications.
bf-update-conventions
by khaki4
스프린트 중 코드 리뷰에서 발견된 반복 패턴을 conventions.md에 반영한다. Convention Guard의 규칙 소스를 축적하여 다음 스프린트의 리뷰 품질을 높인다.
token-cost-optimization
by protick-bjit2019
Use when you need to reduce LLM API token usage or costs — prompt compression, model tiering, caching, context trimming, tool call minimization, output length control, and automated context compression via headroom-ai.
scaffold-skill
by maxoreric
系统初始化脚手架。用于在新项目中克隆和初始化 sop-engine 内核。
personal-context-agent
by Cursedpotential
"Extract personal facts about a user from AI chat transcripts and structure into themed markdown files. Use when (1) Processing Claude, Claude Code, or ChatGPT conversation exports, (2) Building personalized AI context from chat history, (3) Creating context files for Claude Projects, GPTs, or Gems. Optimized for Claude Haiku."
visual-prompt-engine
by Abdullah4AI
"Generate diverse, non-repetitive image prompts powered by real visual references from Dribbble and design platforms. USE WHEN: user wants an image prompt, needs creative visual inspiration, asks for design-informed prompts, wants to avoid repetitive AI image generation, or says 'generate a prompt for an image', 'give me a creative image idea', 'make me a unique visual prompt'. DON'T USE WHEN: user wants to generate the image itself (use an image generation tool), wants to edit an existing image, or needs text-only content. EDGE CASES: 'make me an image' → use image generation tool, then optionally this skill for the prompt. 'improve this image prompt' → this skill. 'I keep getting similar AI images' → this skill (solves repetition)."
prompt-injection-auditor
by screem500
Security audit of LLM system prompts, agent instruction files (SKILL.md, AGENTS.md, CLAUDE.md), and agent configurations against prompt injection attacks. Use when the user wants to (1) audit or harden a system prompt or agent instructions against prompt injection, (2) review an agent skill or system prompt for security weaknesses before publishing, (3) generate a prompt-injection risk report with severity ratings and fixes, (4) run authorized red-team tests against an LLM agent they own or are permitted to test, or (5) check for data-leakage risks such as exposed secrets, weak instruction hierarchy, or missing output constraints.
xiameng-prompt-engineer
by chengbenchao
夏导提示词生成引擎 v3.0:基于《夏导提示词库》标准,将自然语言需求自动转化为 10 维度 × 9 段增强版专业级 AI 绘画提示词。适用于电影分镜、角色设定、场景概念图、 特殊视图(三视图)、多图拼版。当用户要求生成绘图提示词、AI 绘画 prompt、角色设定描述时使用。 触发词:夏导引擎、夏导提示词、xiameng prompt、生成提示词、AI绘画提示词、三视图提示词、夏导快速模式。 支持 S1-S11 十一大场景、三种工作模式(极速/快速/标准)、色温/焦段/光位铁律、细节与质感增强。 v3.0 新增:自然语言快速模式、S11 多图拼版场景、GPT-Image-2 混合标杆库。
cx-summary
by m19803261706
CX 工作流 — 汇总发布与闭环。手动触发或由 cx-exec 全部完成时自动调用。 生成汇总文档，GitHub 同步（基于 config.github_sync）， 智能更新 CLAUDE.md 规范，触发完成通知。 仅在用户明确调用 /cx-summary 或 cx-exec 全部完成时执行。不要自动触发。
intuitive-deep-learning
by ReductTech
在 Electron 内置浏览器中启动并使用本地深度学习互动课程。用于学习或讲解神经元、MLP、激活函数、损失函数、梯度下降、数字图像、卷积、CNN、LeNet 和人脸识别等主题，用于浏览课程图谱、选择学习路径和完成互动练习，也用于用户询问本技能是否有更新、要求检查更新或同步最新版时。
autonomous-claude-sandbox
by WellDunDun
Deploy Claude Code on Cloudflare Sandboxes. Run autonomous AI coding tasks in isolated containers via a simple API.
Onboard Skill
by yousufjoyian
```
viral-video-replicator
by dingtom336-gif
"视频逆向复刻 — 分析参考视频(FFmpeg帧提取+Vision LLM) + 生成复刻Seedance 2.0 Prompt + 4种素材替换模式。支持单个和批量。Use when: '复刻这个视频', '分析爆款视频', 'replicate this video', '视频逆向', '反编译视频', '批量分析视频'. Do NOT use for creating from scratch — use fashion-video-creator instead."
Skill Creator
by jprokay-counterpart
Use when creating new skills, improving existing skills, or when a complex repeatable task emerges that lacks documented expertise. Triggers: "create a skill", "we should document this", "make this reusable", working on complex domain-specific tasks without existing skill coverage (Django ORM, Kubernetes, API design, etc.)
ai-note-2.0
by zzc178178
生成AI笔记（双页签图文笔记）。触发词：AI笔记、AI note、生成笔记、图文笔记、summary note、知识总结、做笔记、整理笔记、AI note、双页签。支持对网页文章生成精美双页签HTML笔记，类似抖音AI笔记样式。包含固定在底部的「文字笔记」和「图解」两个页签，方便滑动后快速切换。
inquirer
by yusenthebot
A collection of common interactive command line user interfaces.. Use when: providing error feedback; asking questions; validating answers. NOT for: web browser UI rendering; server-side HTML generation.
vlog_script_to_speech
by theoneLee
将手写稿转换为适合朗读的口播稿
MindFlow
by chuanhaigogogo
Generate concise mind maps and structured learning notes from AI conversations with Obsidian export support.
daily-meeting-update
by VisualxIntelligence
"Interactive daily standup/meeting update generator. Use when user says 'daily', 'standup', 'scrum update', 'status update', 'what did I do yesterday', 'prepare for meeting', 'morning update', or 'team sync'. Pulls activity from GitHub, Jira, and Claude Code session history. Conducts 4-question interview (yesterday, today, blockers, discussion topics) and generates formatted Markdown update."
debugger
by YPYT1
集合了所有的openclaw,claudecode,open等各个平台所有的skills
daily-digest
by BingqiangZhou
Unified daily digest generator. Monitor tech news, podcasts, and WeChat articles. Trigger: "日报", "科技日报", "播客更新", "微信日报", "全部日报", "digest", "daily report" Do NOT trigger for: creating/editing content, managing accounts.
system-watchdog
by wenjie2024
深度审计 OpenClaw 系统健康状态，分析 Gateway 重启原因、LLM Fallback 记录以及定时任务运行实况。使用中文汇报。