Prompts
Prompt engineering and templates
continuous-mode
by iulspop
Continuously steers the codebase toward the VISION.md. Reads the vision, assesses current state, picks the highest-impact next action, builds with TDD, commits, and repeats. Never stops. Use when you want autonomous, vision-driven development.
ralph-loop
by 101mare
Start an autonomous work loop. Claude keeps working on the given task until it signals completion or the iteration limit is reached. Does NOT handle: prompt building (use ralph-loop-prompt-builder), plan review (use plan-review), code review (use pr-review). Usage: /ralph-loop <task description> Optional: /ralph-loop max=20 <task description> Cancel: delete .claude/ralph-loop.local.md or press Escape/Ctrl+C
call-ai
by choru-k
Run a prompt against external AIs (Codex, Gemini, Claude). Low-level building block for other skills. Use for "ask codex", "ask gemini", "call AI", or when delegating prompts to external models.
auto
by cadrianmae
This skill should be used when the user asks to enable tracking, disable tracking, toggle auto-tracking, turn on/off automatic tracking, pause tracking, resume tracking, or control hooks-based tracking state. Manages .claude/.ref-autotrack marker file to activate/deactivate PostToolUse and SessionEnd hooks for automatic source and prompt tracking.
agnt
by october-academy
Guides users through Agentic30's MUD-style 30-day learning journey. Use when the user asks to start, continue, submit, reset, or check status for Agentic30 quests, onboarding, interview-driven development, or landing deployment.
ralph-loop
by addozhang
Guide OpenClaw agents to execute Ralph Wiggum loops using exec and process tools. Agent orchestrates coding agents (Codex, Claude Code, OpenCode, Goose) with proper TTY support via pty:true. Plans/builds code via PROMPT.md + AGENTS.md, SPECS and IMPLEMENTATION_PLAN.md. Includes PLANNING vs BUILDING modes, backpressure, sandboxing, and completion conditions. Users request loops, agents execute using tools.
cc-dev-skills
by choru-k
Guide for creating effective skills. Use for "create a skill", "new skill", "write a SKILL.md", "update a skill", "skill template", or extending Claude's capabilities with specialized workflows and tools.
ai-model-selector
by jaymay549
"Use when selecting AI models, configuring API parameters, or implementing LLM calls. Covers OpenAI (GPT-5.2, GPT-5.1, GPT-4.1, o3), Anthropic (Claude 4.5), Google (Gemini 2.5/3), DeepSeek (V3.2, R1), and embedding models with specs, gotchas, and code templates."
mixseek-evaluator-config
by drillan
MixSeekの評価設定ファイル（evaluator.toml、judgment.toml）を生成します。「評価設定を作成」「スコアリング設定」「判定設定を作って」「メトリクスを設定」といった依頼で使用してください。Submissionの評価基準と最終判定ロジックを定義します。
thought-mining
by vagus95
思维挖掘助手 - 通过对话帮助用户把脑子里的零散想法倒出来、记录下来、整理成文章。覆盖从思维挖掘到成稿的完整流程。
baoyu-cover-image
by Jackiexiao
Generates article cover images with 5 dimensions (type, palette, rendering, text, mood) combining 9 color palettes and 6 rendering styles. Supports cinematic (2.35:1), widescreen (16:9), and square (1:1) aspects. Use when user asks to "generate cover image", "create article cover", or "make cover".
integration
by cadrianmae
Integrate cadrianmae marketplace plugins into CLAUDE.md. Scans installed plugins and generates documentation section listing available skills and commands. User chooses between user-level (global) or project-level (current project only) integration.
reflection
by Jackiexiao
"Analyzes the conversation and tool usage to propose improvements to skills or store user preferences."
ralph-loop-prompt-builder
by 101mare
Builds effective prompts for the Ralph Loop autonomous work system. Asks clarifying questions about the task, then generates a structured prompt with clear requirements, verification steps, and completion criteria. Does NOT handle: loop execution (use ralph-loop), general prompts (use prompt-builder). Use this BEFORE starting a ralph-loop — it creates the prompt, not the loop itself. Recognizes: "ralph-loop-prompt-builder", "ralph prompt", "build ralph prompt", "prepare a ralph loop task", "help me write a ralph loop prompt", "structure this for ralph-loop"
w3-generate-claude-files
by famtong8-dev
Generate CLAUDE.md, CODEMAP.md, and MEMORY.md for any backend project (Node.js, Python, Go, etc.) using prompts and manual analysis. Uses Claude AI to help you document your project constitution, codebase architecture, and context knowledge. Language-agnostic, prompt-based approach - focused on accurate, personalized documentation through interactive analysis.
migration-skill
by PloLink
从 Cursor 开发模式迁移到 Claude Code 开发模式的完整指南。当项目需要从 Cursor AI IDE 切换到 Claude Code CLI 时使用此技能，包括配置迁移、文档结构转换、工作流适配等。
claude-code-wrapper
by xdrshjr
Simplified Claude Code CLI wrapper that automatically handles environment variables from ~/.claude/config.json. Use when the user wants to run Claude Code commands without manually exporting ANTHROPIC_AUTH_TOKEN, ANTHROPIC_BASE_URL, and API_TIMEOUT_MS. Eliminates the "Not logged in" error when calling Claude Code through exec tools. Perfect for automating Claude Code tasks in OpenClaw sessions.
claude-code-guide
by ya-luotao
Use this agent when the user asks questions ("Can Claude...", "Does Claude...", "How do I...") about: (1) Claude Code (the CLI tool) - features, hooks, slash commands, MCP servers, settings, IDE integrations, keyboard shortcuts; (2) Claude Agent SDK - building custom agents; (3) Claude API (formerly Anthropic API) - API usage, tool use, Anthropic SDK usage. IMPORTANT: Before spawning a new agent, check if there is already a running or recently completed claude-code-guide agent that you can resume using the "resume" parameter.
add-ext-skill
by unicorn-plugins
확장 스킬 추가 (외부 SKILL.md 동적 로드)
product-marketing-context
by Jackiexiao
"Create/update a reusable product marketing context document to align positioning and messaging."
docs-workflow
by fefogarcia
Four slash commands for documentation lifecycle: /docs, /docs-init, /docs-update, /docs-claude. Create, maintain, and audit CLAUDE.md, README.md, and docs/ structure with smart templates. Use when: starting new projects, maintaining documentation, auditing docs for staleness, or ensuring CLAUDE.md matches project state.
send
by cadrianmae
Send context to parent, child, or sibling session before switching
pyspider-dev
by Within-7
"Production-ready PySpider web crawler development toolkit. Use when: (1) Creating new PySpider crawlers for web scraping, (2) Implementing complex crawling with proxies and retries, (3) Handling pagination, authentication, and data extraction, (4) Building enterprise-grade crawlers with error handling. Provides templates, patterns, and best practices for reliable scraping."
prompt-lookup
by baotoq
Activates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.