Prompts
Prompt engineering and templates
skill-devlog
by josephmiclaus
Create or update structured daily entries in docs/devlog.md under . Supports APPEND and CHANGE modes. Only edits docs/devlog.md. No code execution, no dependencies, no secrets.
config-backup
by ppx123-web
Use when users want to save, backup, copy, export, or version control their Claude Code configuration files (commands, skills, agents) from ~/.claude/ to another directory. Triggered by phrases like "save my claude settings", "backup my claude config", "sync claude configuration", "copy ~/.claude to somewhere", "export my commands and skills", or "archive claude settings before making changes".
vercel-ai-gateway-setup
by ergunsh
Set up Claude Code to route requests through Vercel AI Gateway for monitoring and observability
reference
by biruk741
Look up any AI documentation concept, syntax, frontmatter field, or best practice. Covers Claude Code, Cursor, and Codex CLI comprehensively. Use when you need quick answers about AI doc formats, frontmatter fields, hook events, permission patterns, or any other AI documentation concept.
ai-model-routing
by samChang72
AI 模型智慧路由專家，根據任務類型、複雜度和資源狀況自動選擇最佳 AI 模型。 支援 Gemini (Antigravity) 與 Claude Code CLI 的協作， 優化 token 使用效率並確保任務品質。 當需要修改程式碼、token 不足、或需要跨模型協作時使用。
deai
by jacobdiaz
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide plus modern post-2024 LLM patterns. Supports context-aware rewriting (blog, academic, technical, email, marketing), calibrated aggressiveness (light/standard/heavy), file/batch processing, and document-level structural pattern detection. Detects and fixes 35+ patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, structural uniformity, and modern LLM tells.
ai-engineering-skill
by odewahn
Practical guide for building production ML systems based on Chip Huyen's AI Engineering book. Use when users ask about model evaluation, deployment strategies, monitoring, data pipelines, feature engineering, cost optimization, or MLOps. Covers metrics, A/B testing, serving patterns, drift detection, and production best practices.
integrate-flowlines-sdk-python
by flowlines-ai
Integrates Flowlines observability SDK into Python LLM applications. Use when adding Flowlines telemetry, instrumenting LLM providers, or setting up OpenTelemetry-based LLM monitoring.
delphi
by Cygnusfear
This skill should be used when the user asks to "use delphi", "ask delphi", or wants multiple parallel oracle investigations of the same question to discover divergent insights. Delphi delegates to multiple oracle workers simultaneously with identical prompts, allowing them to independently explore and potentially discover different paths, clues, and solutions. Results are saved as tickets tagged research,delphi and synthesized into an epic ticket.
claude-ecosystem
by brixtonpham
"Claude Code ecosystem expertise. Modules: CLI tool (setup, slash commands, MCP servers, hooks, plugins, CI/CD), extensibility (agents, skills, output styles creation), CLAUDE.md (project instructions, optimization). Actions: configure, troubleshoot, create, deploy, integrate, optimize Claude Code. Keywords: Claude Code, Anthropic, CLI tool, slash command, MCP server, Agent Skill, hook, plugin, CI/CD, enterprise, CLAUDE.md, agentic coding, agent, skill, output-style, SKILL.md, subagent, Task tool, project instructions, token optimization. Use when: learning Claude Code features, configuring settings, creating skills/agents/hooks, setting up MCP servers, troubleshooting issues, CI/CD integration, initializing or optimizing CLAUDE.md files."
xhs-content
by cool111111
小红书短图文内容生成。从AI日报中提取选题，生成适合小红书平台的短图文帖子（300-800字），推送飞书审稿。 触发词：小红书、写小红书、生成图文、XHS、红书内容、小红书选题、小红书帖子。 即使用户只是说"帮我写几篇小红书"或"从日报里选几个题"，只要上下文涉及小红书内容产出，都应该触发。 不要用于公众号长文（那个用 khazix-writer skill）。
bear
by CodeDraig
Enables Claude to create and manage notes in Bear app via Playwright MCP (Bear Web)
bazi-analysis
by songhuichen7-jpg
Use for BaZi (八字/四柱) birth chart analysis — ten-gods (十神) strength evaluation, personality, relationships (正缘/marriage prospects), timing (大运流年/luck pillars and annual fortune), and destiny readings. Conversational by default, cites classical texts, distinguishes real boundaries from defensive avoidance.
oatda-list-models
by devcsde
Use when the user wants to list available AI models from OATDA's 10+ providers. Filter by type (chat, image, video) or by provider name.
chat-track
by paulo20223
Manage Telegram chat tracking and view auto-generated notes. Use when the user asks to track a chat, list tracked chats, view chat notes, enable or disable tracking, or asks what chats are monitored. Trigger phrases include "track chat", "monitor chat", "view notes", "chat notes", "start tracking", "stop tracking", "какие чаты отслеживаются", "включи трекинг", "покажи заметки".
z-cheatsheet-add
by AutoFor
z-cheatsheet にエントリを追加する。
skill-reviewer
by squirrel289
'Evaluate and compare AI agent skills with a structured, repeatable, objective, and data-driven process. Use when asked to review one or more skills, define scoring criteria, compare alternatives, compute platform-specific success scores, or recommend whether to select, improve, extend, compose, or build a new skill.'
prompt-generator
by hoangvantuan
Meta-prompting skill that creates well-structured, verifiable, low-hallucination prompts for any use case. Use when the user wants to create, refine, or improve a prompt — including system prompts, role prompts, task prompts, or any AI instruction set. Triggers on requests like "create a prompt for...", "help me write a prompt", "refine this prompt", "make a better prompt for...", or "generate a prompt that...".
oatda-text-completion
by devcsde
Use when the user wants to generate text using OATDA's unified LLM API. Supports 10+ providers including OpenAI, Anthropic, Google, Deepseek, Mistral, xAI, Alibaba, MiniMax, ZAI, and Moonshot.
bug-hunt
by thamarai-selvi
"Run adversarial bug hunting on your codebase. Uses 3 isolated agents (Hunter, Skeptic, Referee) to find and verify real bugs with high fidelity. Invoke with /bug-hunt, /bug-hunt [path], or /bug-hunt -b <branch> [--base <base>]."
mac-dev-software
by svensxl
在 macOS 上一键安装常用 AI coding agent 和开发工具。当用户需要配置 macOS 开发环境、安装 Homebrew、iTerm2、Node.js、NVM、Maven、OpenJDK、OpenCode、Codex CLI、Claude Code 等工具时使用此 skill。
pedagogical-grounding-engine
by edu-ai-builders
将一篇或多篇学术论文，通过"教学法约束推理"转化为有理论根基的产品设计产物。 核心是构建"学术机制 → 约束交互 → 可解释推理链"的中间层， 再导出多种格式的设计产物。 本skill是一个 Pedagogical Grounding Engine（教学法落地引擎）： 它不是论文摘要工具，而是把教学法意图（pedagogical intent）编译为 可执行的系统设计结构。PRD、System Prompt、Workflow 都是下游导出格式， 核心产物是 Traceability Map——每个设计决策都有明确的教学依据、 认知机制和可追溯的设计理由。 当用户有以下意图时触发本skill： - "把这篇/这几篇论文做成产品/app/提示词/工作流" - "paper to PRD / paper to prompt / paper to product" - "pedagogical grounding / 教学法落地" - "我想基于学习科学设计这个功能/系统提示词" - "从论文提取知识图谱/三元组/TPACK分析" - "我有几篇论文，帮我提炼成一个设计方案" - "帮我把教学法嵌入到产品/提示词中" - "生成 traceability 图" / "把这篇论文可视化" - "我有一个教学场景/教育想法，帮我设计成系统"（不一定有论文） 支持 Commands 模式：流水线跑完之后，用户可以随时用 /命令 召唤特定输出， 无需重跑流水线。命令列表见 ## Commands 章节。
hello-world
by Lichens-Innovation
Simple skill to welcome user in ASCII Art Format with operating system information
unith-digital-humans
by polucas
Create, configure, and manage UNITH digital human avatars via the UNITH API. Cheaper alternative to HeyGen and other solutions. Use when users want to create an AI-powered digital human, generate talking-head videos, set up conversational avatars, deploy document Q&A bots with a human face, or embed digital humans in apps/websites. Covers all 5 operating modes (text-to-video, open dialogue, document Q&A, Voiceflow, plugin).