AI & ML

nvidia Bundle

Use when designing, configuring, running, or troubleshooting NeMo Agent Toolkit evaluations, datasets, evaluator selection, ATIF surfaces, quality gates, custom evaluators, and `nat eval`.

nvidia Bundle

Use when installing or configuring NVIDIA NeMo Agent Toolkit, verifying the `nat` CLI, setting up optional extras, or creating a first hello-world workflow.

nvidia Bundle

Use when serving NeMo Agent Toolkit workflows, exposing workflows through FastAPI, configuring MCP clients or servers, or troubleshooting transport and server setup.

nvidia Bundle

Use when authoring, registering, composing, or testing custom NeMo Agent Toolkit tools, functions, function groups, Python components, custom agents, custom evaluators, or advanced extension patterns.

codethread Bundle

Mandatory when configuring any Claude Code specifics - hooks, skills, plugins, MCP servers, slash commands, settings, agents, or any Claude Code feature. Provides opinionated best practices

cskwork Skill

Run the llm-review .task plan, sequential review, implementation, and review workflow for an explicitly requested development task.

cskwork Skill

Run schema-constrained Codex review of the llm-review .task plan or implementation after the configured preceding reviews.

cskwork Skill

Review the llm-review .task plan or implementation for correctness, security, and test coverage before the configured next reviewer.

cskwork Skill

Implement the approved llm-review plan in .task/plan-refined.json and record changes, tests, and blockers in impl-result.json.

wxy Skill

自动汇总当前分支相对主分支的提交摘要，用于 PR/说明对齐。

wxy Skill

会话输出安全控制。防止超长输出导致失败，必要时分段写入文件。

wxy Skill

执行前检查清单。每次回复前必须自动进行的自我检查，确保没有遗漏强制规则。

wxy Skill

校验“说明内容 ↔ 实际变更文件”的一致性，避免描述遗漏或偏差。

jaccen Skill

---

ancplua Bundle

Example skill in the full-plugin fixture. Use when cc-plugin-eval tests need a valid skill alongside agents, hooks, MCP, LSP, monitors, themes, and userConfig.

throokie Bundle

智能提示词专家 - 根据上下文和对话环境动态生成最优提示词

haru-llc Skill

実装計画作成スキル。マルチステップ開発タスク向けの詳細な計画を生成する。

haru-llc Skill

計画実行スキル。事前作成された計画を別セッションで実装する構造化アプローチ。

haru-llc Skill

進捗報告書を作成する。定期進捗報告、マイルストーンレビュー時に使う。

haru-llc Skill

Git Worktreeスキル。共有リポジトリ内で分離されたワークスペースを作成し、並列ブランチ作業を可能に。

ccf-claude-code-ccf-marketplace Bundle

Create high-quality MCP servers for LLM tool integration.

ccf-claude-code-ccf-marketplace Skill

Implement comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking. Use when testing LLM performance, measuring AI application quality, or establishing evaluation frameworks.

ccf-claude-code-ccf-marketplace Skill

Select and optimize embedding models for semantic search and RAG applications. Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality for specific domains.

wolverin0 Skill

Generate a detailed AI fix prompt for a specific audit finding. Use when the user asks "/audit-fix F-X.Y" or "give me the fix prompt for finding N." Produces a complete prompt the user can hand to a fresh AI session (Claude Code, Cursor, ChatGPT) to remediate the finding with verification and rollback steps.

technicalpickles Skill

Resume work from a parked handoff

technicalpickles Bundle

Harness Binary Spelunking

abdullahmalik17 Bundle

Build production RAG systems with semantic chunking, incremental indexing, and filtered retrieval. Use when implementing document ingestion pipelines, vector search with Qdrant, or context-aware retrieval. Covers chunking strategies, change detection, payload indexing, and context expansion. NOT when doing simple similarity search without production requirements.

pr-pm Bundle

Use when creating, improving, or troubleshooting Claude Code subagents. Expert guidance on agent design, system prompts, tool access, model selection, and best practices for building specialized AI assistants.

davidme6-self-learning-skill Bundle

Self-Improving Agent v2.0 - 自我进化智能体

treasure-data Bundle

LLM Workflow

treasure-data Skill

Use this skill at the end of a Google Slides deck-creation flow run by the same agent in the current session, before reporting the finished URL to the user. Catches the failure modes a creating agent reliably misses — surviving placeholder text, unswapped template imagery, layout overflow, hidden-original leaks. Trigger when the parent google-slides workflow reaches Step 9 / QA, or when the user asks "is this deck ready?" / "QA my deck" about a deck this agent itself just edited. Do NOT trigger for reviewing decks the agent did not create — without the Step 5 pre-fill snapshot the leak check silently degrades and the result is misleading.

dolsoon Bundle

Handle template comparison and optimization workflows including backup detection, smart merging, and update completion reporting. Enhanced with Context7 MCP for up-to-date documentation.

faberlens Bundle

Stop waiting for prompts. Keep working.

coderabbitai Skill

Browser automation CLI for AI agents. Use when the user needs to interact with websites, verify dev server output, test web apps, navigate pages, fill forms, click buttons, take screenshots, extract data, or automate any browser task. Also triggers when a dev server starts so you can verify it visually.

nickolaslin33 Bundle

人與 AI agent 工作一整天之後要離開、或隔了幾小時到一個週末回來接手時，用這套流程留下與讀取「重建材料」。收工分四個階段——人離開工作畫面憑記憶寫、agent 校驗、人修正、整理任務狀態；回來時照固定順序讀出來。它解的不是記憶問題：在這個時間尺度上恢復成本十幾秒就飽和，回來時發生的幾乎都是重建，真正的病是理解債——在尚未真正理解前就讓任務往下推進，症狀看起來像忘了，實際是當時就沒形成足夠的理解。觸發情境：使用者說「我回來了」「繼續某某專案」「上次做到哪」「我忘記做到哪了」「重新進入狀況」「哪個專案還沒做完」，或說「收工」「今天到這」「先記一下進度」「交接一下」「幫我留個紀錄」；也包含任何檢視完 agent 的產出準備離開、或隔了一段時間要接回某個專案的時刻。使用者不必說出 reentry 這個詞——只要他們在問「我上次在幹嘛」或正要結束一段工作，就用這個 skill。Also triggers on — resume work, pick up where I left off, where was I, wrap up, end of session, handoff notes, context recovery, what did I do last time.

nickolaslin33 Bundle

寫任何會離開這台機器的東西時，範例素材一律用虛構但一樣具體的名字，不要把手邊的真實專案名、家目錄路徑、email、GitHub 帳號、公司內部系統代號寫進去。用在撰寫 skill（SKILL.md、references/、evals/）、README、公開 repo 的文件、部落格、對外的技術報告、要貼到 issue 或聊天室給別人看的程式碼片段與終端機輸出。特別是被 skill-creator 觸發、或正在寫任何要 push 到公開 repo 的文件時，務必套用——那個情境最容易把真實資訊寫進去，因為 Agent 的 context 裡剛好有當前工作目錄、工作區裡其他專案的名字與這段對話提到的真實專案，需要舉例時它們是最現成的材料。觸發詞：「去識別化」「不要用我的專案當範例」「這要公開」「要 push 到公開 repo」「寫 skill」「寫 README」「整理成對外文件」「匿名化」「這段可以貼出去嗎」「幫我看有沒有夾帶個資」。反過來說，只在本機的筆記、私有 repo 的內部文件、commit 訊息、程式碼註解不適用——那些地方寫真實名稱才是對的。