1. 2545Project OrchestratorFreddy-CHEN-ux/00_claude_skillsOrchestrate multi-phase project implementation using background Claude agents for isolated development, review, and integration.8
  2. 2546Remote Code AgentsFreddy-CHEN-ux/00_claude_skillsDelegate tasks to remote Claude Code agent containers for parallel execution, long-running analysis, or resource-intensive operations.8
  3. 2547Router EngineeringFreddy-CHEN-ux/00_claude_skillsMaster orchestration for routing technical problems through 32 engineering skills - AI/ML, software development, data, APIs, and Claude Code framework8
  4. 2548Skill Composerbolasblack/claude-skillsUse when the user asks to create a new Agent Skill, revise an existing skill's authored behavior or activation, audit a complete skill package, design its eval suite, port it across agent harnesses, or prepare and validate its release. When a generic skill-authoring helper overlaps, this skill owns scope, structure, portability, validation, and release history. If the work neither changes nor audits an Agent Skill package, do not load this skill.5
  5. 2549Skilzy CreatorFreddy-CHEN-ux/00_claude_skillsComprehensive toolset for creating, converting, validating, and packaging Skilzy-compliant AI agent skills. This skill should be used when users want to initialize a new skill, convert a Claude skill archive to the Skilzy format, validate an existing skill's structure and compliance, or package a skill for distribution to the Skilzy registry.8
  6. 2550Statechartbolasblack/claude-skillsCheck a TLA+ model of a system that moves between states, such as a product flow, a protocol, or a service, and draw it as a Mermaid statechart. The report gives a shortest path for each rule that breaks. Use for /statechart, or when a task needs a TLA+ model checked or a model's bug shown as a path through states.5
  7. 2551Subagent Driven DevelopmentFreddy-CHEN-ux/00_claude_skillsUse when decomposing complex work. Dispatch fresh subagent per task, review between tasks. Flow: Load plan → Dispatch task → Review output → Apply feedback → Mark complete → Next task. No skipping reviews, no parallel dispatch.8
  8. 2552Subagent WorkflowFreddy-CHEN-ux/00_claude_skillsUse when orchestrating subagents for implementation. Covers when to dispatch vs recall, how to scope tasks, session management, and handoff patterns. Load this before working with any subagent.8
  9. 2553System Architectcatch-the-wave/fullstack-ios-claude-skillsBackend architecture, API design, and ML/AI systems expertise. Use for API design, data flows, ML pipelines, embeddings, vector databases, or architectural decisions. Researches best practices via Perplexity when uncertain.5
  10. 2554Understanding MistakesFreddy-CHEN-ux/00_claude_skillsUse when user says something went wrong, AI did Y instead of X, or wants to understand why a deviation happened. Manually invoked post-mortem analysis.8
  11. 2555Update Langfuse Staging Server PromptFreddy-CHEN-ux/00_claude_skillsPush prompt updates to Langfuse (staging or production). Defaults to STAGING (safe). Production requires explicit --production flag + confirmation. NEVER assigns labels (human-in-the-loop safety).8
  12. 2556ActingGAGAA-HUE/ASHUO-claude-skills角色表演系统——专为 AI 视频生成(Seedance 2.0)设计的角色行为/表演指导系统。 核心理念:表演是压力下的行为,而非情绪展示。将角色的目标、障碍、策略、 节拍变化、潜台词分解为可观察的身体语言和微表情。擅长:角色动机设计、 情绪节奏控制、非语言行为描述、反应时序、身体状态锚定。触发词:角色表演、 演技设计、角色行为、表演指导、acting、character performance、emotional beats。5
  13. 2557Art Asset DesignerGAGAA-HUE/ASHUO-claude-skills美术资产设计师——专为剧本/故事内容拆解人物与场景美术资产,并为每个资产生成可直接用于 AI 图像生成工具的提示词。当用户提到剧本、故事、脚本、分镜、需要美术资产、出图提示词、角色设计、场景设计等内容时,立即触发此技能。即使用户只是说"帮我拆一下剧本里的角色"或"我需要生成场景图",也应立即以美术资产设计师身份介入。覆盖从剧本解析 → 资产清单整理 → 风格确认(必须向用户询问,禁止自行推断)→ 分类提示词生成的完整流程。5
  14. 2558Ccr Modelsnpow/claude-skillsUse when the user asks "what models are available", "list models", "show models", or wants to see which LLM models are accessible via claude-code-router.5
  15. 2559Cinedance HiggsfieldGAGAA-HUE/ASHUO-claude-skills电影级视频提示词生成器,专为 Seedance 2.0 设计。使用 4D 方法论(解构-诊断-开发-交付)生成高预算电影质感的 AI 视频提示词。擅长:空间布局控制、光学镜头设计(FOV度数系统)、物理真实性、多镜头剪辑序列、角色/场景参考管理、光影锁定。支持单镜头长镜头或多镜头剪辑序列,任意时长。触发词:Seedance、电影级视频、分镜脚本、镜头设计、视频提示词、场景调度、光学设计、多镜头序列。5
  16. 2560Claude HandoffGAGAA-HUE/ASHUO-claude-skillsHand the current conversation off to a fresh background agent that picks up the work immediately.5
  17. 2561Design an InterfaceGAGAA-HUE/ASHUO-claude-skillsGenerate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions "design it twice".5
  18. 2562Ice Cave ExplorerGAGAA-HUE/ASHUO-claude-skills冰洞探险视频剧本 skill —— 为 AI 视频生成工作流产出剧本层资产(亮点设计、场景设定、分镜脚本)。每个产出物对应 120 秒横屏冰洞探险短片的一个环节。不产 AI 工具提示词、不产操作手册。触发词:冰洞、探险视频、冰洞剧本、ice cave。5
  19. 2563Magic Fetchnpow/claude-skillsLogs capability gaps as they arise during a session — what Claude couldn't do and exactly what tool, access, or data would have made it possible. Use when the user wants to capture integration needs, track where Claude hits walls, build a roadmap from blocked requests, log capability gaps, or says "remember what you couldn't do". Auto-activates when Claude is about to say it cannot access, fetch, or perform something.5
  20. 2564Ml Pipeline Reportnpow/claude-skillsUse when checking ML pipeline health, Metaflow flow status, failed runs, or generating an ML pipeline report. Trigger phrases: metaflow status, ML pipeline health, flow runs, metaflow report, are my flows running, ML pipeline check.5
  21. 2565Prompt Clashnpow/claude-skillsAll-in-one prompt engineering competition toolkit — attack generation, defense hardening, real-time analysis, and pattern reference for AI security tournaments like Clash of Prompts. Use when preparing for or competing in prompt engineering competitions, red-teaming LLMs, hardening system prompts, analyzing prompt injection techniques, or practicing timed prompt challenges. Trigger phrases include "prompt clash", "prompt competition", "prompt tournament", "red team this prompt", "harden this prompt", "attack this prompt", "defend this prompt", "prompt injection", "prompt security", "clash of p5
  22. 2566Prompt Clash Ensemblenpow/claude-skillsFans out 4 parallel prompt-clash defend agents at staggered time budgets (1min, 2min, 3min, 4min), then ensembles/synthesizes the best elements of each into a single hardened prompt. Use when the user wants the strongest possible defense prompt and has ~5 minutes. Trigger phrases include "prompt clash ensemble", "ensemble defense", "fan out prompt clash", "best defense possible", "ensemble prompt", "multi-budget defense".5
  23. 2567Retrospectnpow/claude-skillsSession retrospective that scans the current conversation for agent failures, user corrections, and alignment gaps, then proposes patches to CLAUDE.md or skills. Use when: after a difficult session, after repeated corrections, to improve agent behavior, to analyze what went wrong, to close the learning loop after autopilot/fix-pr. Trigger phrases: "retrospect", "what went wrong", "session retro", "learn from this", "what should we improve", "session review", "capture learnings", "post-mortem this session".5
  24. 2568Seedance DirectorGAGAA-HUE/ASHUO-claude-skills电影级视频分镜头提示词生成器,专为 Seedance 2.x 系列模型设计。当用户需要生成视频、制作分镜脚本、描述影像画面、进行镜头调度、输出 AI 视频提示词(prompt)时触发。默认输出适合直接复制给模型的轻量成品提示词,优先贴近长镜头、多分镜、蒙太奇三类实用模板。动作戏默认采用事件链长镜头逻辑;变身、异变、强烈招式前摇等场景采用阶段推进逻辑。多分镜默认强制机位四维轮换(景别 / 角度 / 视点 / 运动方式)以避免构图单调,并严格守恒 180 度轴线以避免越轴。仅当用户明确要求工业版、完整版、分镜表或后期说明时,才输出重型结构化版本。5
  25. 2569Suno PrompterGAGAA-HUE/ASHUO-claude-skillsSuno v5.5 Pro 音乐提示词 + 歌词生成专家——将用户提供的任意简单需求(情绪、场景、歌词方向、用途)转化为符合 Suno v5/v5.5 规则的完整方案:Style Prompt(英文风格提示词)+ 专业歌词(含韵脚/节奏/音节控制)+ 负向约束。当用户提到 Suno、AI 音乐生成、帮我写提示词、帮我生成歌曲、我想用 Suno 做一首歌、Suno prompt 怎么写、风格提示词、歌词、作词、帮我写歌、AI 作曲等任何场景时,立即触发此 skill,主动介入并输出完整可粘贴方案。即使用户只是说"我想做一首悲伤的歌"也应直接生成,不要先问用户要不要使用此技能。5
  26. 2570Team Temporalnpow/claude-skillsUse when coordinating multiple agents on a staged pipeline — plan → PRD → exec → verify → fix — via a durable Temporal-backed workflow that survives session crashes. Trigger phrases include "team temporal", "sagaflow team", "durable team", "temporal-backed agent team". File-based state, independent critic/verifier gates, two-stage review on every source modification, honest termination labels. Fire-and-forget while you do other work.5
  27. 2571Update Stop Gatenpow/claude-skillsUse when the user describes a bad Claude behavior pattern to add as a stop-gate rule. Handles updating both autonomy-rules.md and the classifier prompt in stop-gate.py. Trigger phrases: update stop gate, add stop gate rule, new autonomy rule, add stall class, update autonomy rules, add behavior to stop gate.5
  28. 2572GuashuaiDENGYUFAN0/guashuai-junshi挂帅模式(orchestrator):主会话模型(STRONG)挂帅坐镇,把任务拆成边界清晰的子任务,点将派给子代理模型(CHEAP,worker 子代理)并行执行,帅只做拆解、派兵、验收、整合,省下执行环节的贵模型额度。适合多文件批量修改、多线并行的大任务。MANUAL-INVOKE ONLY — 平时绝不自动触发;即使用户说"任务很大""帮我并行做""拆一下任务"也不激活(就算 CLAUDE.md 里配置了开工前弹提示规则,那也只是建议弹窗,不算触发本模式)。仅当用户显式输入 /guashuai,或消息以「挂帅」开头时才进入。本轮任务打完即收兵;用户说「收兵」「退出」立即结束。1
  29. 2573JunshiDENGYUFAN0/guashuai-junshi军师模式(advisor):主会话模型(CHEAP)做主帅在前线干活,军师子代理(STRONG)只在架构取舍、疑难 bug 定位、方案评审等关键决策点被请出来问一次,日常执行不动用,最大限度省贵模型额度。适合日常开发的省钱默认姿势。MANUAL-INVOKE ONLY — 平时绝不自动触发;即使用户说"帮我省点额度""这个问题很难"也不激活(就算 CLAUDE.md 里配置了开工前弹提示规则,那也只是建议弹窗,不算触发本模式)。仅当用户显式输入 /junshi,或消息以「军师」开头时才进入。持续生效直到用户说「散帐」「退出」。1
  30. 2574AI Agent BuilderDreydrey9000/claude-skills9-step framework for building production-ready AI agents from scratch. Use when users want to create AI agents, automate with AI, or build multi-agent systems. Keywords: agent, ai, automation, llm, workflow, tool use, multi-agent5
  31. 2575AI Avatar CreatorDreydrey9000/claude-skillsBuild photorealistic AI avatars for content creation using a proven 6-step workflow. Use this skill when users want to create consistent AI characters/avatars for social media content, UGC videos, digital spokespersons, or talking head videos. Triggers include requests to create AI avatars, build consistent AI characters, generate talking AI videos, make virtual influencers, or produce AI-generated spokesperson content. Covers the complete pipeline from base image generation through lip-synced video output.5
  32. 2576Claude Code GuideDreydrey9000/claude-skillsMaster guide for using Claude Code effectively. Includes configuration templates, prompting strategies \\"Thinking\\" keywords, debugging techniques, and best practices for interacting wit...5
  33. 2577Claude Code Starter PackDreydrey9000/claude-skillsThe definitive Claude Code setup system. Use when onboarding anyone to Claude Code, setting up a new machine, auditing your configuration, or answering "what should I install?" Covers ALL 14 hook events, 3 hook types, agent teams, worktree isolation, MCP servers, skills, and proven patterns. Triggers on "starter pack", "setup Claude Code", "onboard", "what hooks do I need", "get started", "new machine setup", "Claude Code features".5
  34. 2578MCP IntegrationDreydrey9000/claude-skillsThis skill should be used when the user asks to "add MCP server", "integrate MCP", "configure MCP in plugin", "use .mcp.json", "set up Model Context Protocol", "connect external service", mentions "${CLAUDE_PLUGIN_ROOT} with MCP", or discusses MCP server types (SSE, stdio, HTTP, WebSocket). Provides comprehensive guidance for integrating Model Context Protocol servers into Claude Code plugins for external tool and service integration.5
  35. 2579Multi Agent PatternsDreydrey9000/claude-skillsMaster orchestrator, peer-to-peer, and hierarchical multi-agent architectures5
  36. 2580Prompt EngineeringDreydrey9000/claude-skillsExpert guide on prompt engineering patterns, best practices, and optimization techniques. Use when user wants to improve prompts, learn prompting strategies, or debug agent behavior.5
  37. 2581Crosscheckmoveju112/skill_verifyClaude-hosted Claude↔Codex crosschecks of improvement ideas, plans, and completed changes. Automatically selects independent proposals, draft critique, or mutual review and allows candid rejection. Use for explicit requests such as "크로스체크", "코덱스랑 핑퐁", "둘이 의견 취합해", "cross-check with codex", or "/crosscheck". Discussing or editing this skill does not authorize another model call.1
  38. 2582Verifymoveju112/skill_verifyEvidence-based local verification and explicitly requested Claude↔Codex crosschecks of improvement ideas, plans, and changes. Automatically selects independent proposals, draft critique, or mutual review, with candid recommendations and minimal exchanges. Shared by Claude Code and Codex. Generic verification stays local; discussing this skill is not permission to call another model. Direct multi-angle unit-test requests use the unit-test skill.1
  39. 2583Client WorkspaceCalebAnderson-Coder/ai-agency-os-starterCreate and maintain an isolated workspace per client, with a client CLAUDE.md holding brand voice, rules, banned claims, contacts, access map, and logs. Use when adding a client, switching clients, or when Claude must "act as" a specific client.1
  40. 2584Close KitchenTalonT-Org/AutoSkillitClose the AutoSkillit kitchen — hides kitchen MCP tools for this session.5
  41. 2585Open KitchenTalonT-Org/AutoSkillitOpen the AutoSkillit kitchen — reveals all kitchen MCP tools for this session. Human-only entry point.5
  42. 2586Planner Elaborate AssignmentsTalonT-Org/AutoSkillitElaborate all assignments for a target phase via parallel L0 subagents. One invocation per phase; spawns one L0 per assignment concurrently. (Pass 2 loop body)5
  43. 2587Planner Elaborate WpsTalonT-Org/AutoSkillitElaborate all work packages for a target phase via parallel L0 subagents. One invocation per phase; spawns one L0 per WP concurrently. (Pass 3 loop body)5
  44. 2588Planner Refine AssignmentsTalonT-Org/AutoSkillitRefine elaborated assignments for a single phase via parallel L0 subagents (L1+L0 pattern), using per-phase context file with peer_summaries for cross-phase visibility5
  45. 2589Planner Refine PhasesTalonT-Org/AutoSkillitRefine elaborated phases with cross-phase visibility via parallel L0 subagents (L1+L0 pattern)5
  46. 2590Planner Refine WpsTalonT-Org/AutoSkillitRefine elaborated work packages with cross-phase visibility via per-phase L0 subagents (L1+L0 pattern)5
  47. 2591Resolve Design ReviewTalonT-Org/AutoSkillitTriage STOP verdict findings from review-design, classifying each as ADDRESSABLE/STRUCTURAL/DISCUSS using parallel subagents. If any are ADDRESSABLE or DISCUSS, generate revision_guidance and emit resolution=revised. If all are STRUCTURAL, emit resolution=failed for terminal stop.5
  48. 2592MCP Builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).