发现 Skills

Long-running coding agent workflows (background + workdir), with tmux for interactive sessions.

Facilitates structured ideation before any architecture or code is written. Use when starting a new feature, product, or system to explore the solution space before committing.

Mark a plan as completed and archive it. Use this when you've finished implementing all items in a plan document.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Diagnose Claude Code environment health: context, description obesity, trigger collisions, hooks, MCP, CLAUDE.md/memory. Graded + observational areas with actionable levers. Use for setup audits or when Claude feels slow.

This skill should be used when the user wants to interact with NotebookLM notebooks via Claude Code's Chrome integration. Trigger phrases: "Query my NotebookLM", "Ask my notebook about X", "query [id] about X", "list my notebooks", "add notebook URL", "show notebook details", "search notebooks for X", "Check my docs", "what does my [topic] notebook say about", "remove notebook", "delete notebook", "disable notebook", "enable notebook". Also triggers when user: (1) mentions NotebookLM explicitly, (2) shares NotebookLM URL (https://notebooklm.google.com/notebook/...). Do NOT use for: general web searches, local file reading, or non-NotebookLM documentation queries. Requires: claude --chrome with claude-in-chrome MCP.

A Unity performance optimization handbook covering CPU, GPU, Memory, Rendering and Physics optimizations. Use the Profiler to analyze bottlenecks and apply targeted optimization strategies. When to use: low game FPS, stuttering, high memory usage, slow loading, frequent GC, use the Profiler, optimize rendering, optimize physics, reduce draw calls. CRITICAL: Always measure with the Profiler first — do not guess the bottleneck.

The 'Brain' of the project: manages state (Context), accumulates knowledge (Patterns + Lessons), and evolves rules.

SEO fundamentals, E-E-A-T, Core Web Vitals, and Google algorithm principles.

Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc.

撰写国家社会科学基金（NSSFC）年度项目申请书，依据 2025 年全国哲学社会科学工作办公室官方模板，覆盖重点项目（A）、一般项目（B）、青年项目（C）、西部项目（X）四类。提供分阶段写作工作流：数据表、选题说明（300字）、选题依据、研究内容、创新之处、研究基础、经费概算、活页匿名化等模块；内置社科文献检索策略与质量自查清单。工具中性，适配 Claude Code / Cursor / Codex / OpenClaw / Gemini CLI。触发词：国家社会科学基金、国社科、社科基金、NSSFC、哲学社会科学、重点项目、一般项目、青年项目、西部项目、社科申报、活页

Uses Exa API for intelligent web searches instead of default WebFetch. Provides up-to-date information with semantic search capabilities.

Gemini CLI for one-shot Q&A, summaries, and generation.

Calculate driving times, distances, and routes using the Google Maps Directions API with real-time traffic data. Use when the user asks for drive time, traffic conditions, commute estimates, or routing between locations. Requires the GOOGLE_PLACES_API_KEY environment variable (the same key works for the Directions API).

Perform language- and framework-specific security best-practice reviews and suggest improvements. Trigger only when the user explicitly requests security best-practices guidance, a security review/report, or secure-by-default coding help. Trigger only for supported languages (Python, JavaScript/TypeScript, Go). Do not trigger for general code review, debugging, or non-security tasks.

Guide for implementing Google Gemini API document processing - analyze PDFs with native vision to extract text, images, diagrams, charts, and tables. Use when processing documents, extracting structured data, summarizing PDFs, answering questions about document content, or converting documents to structured formats. (project)

Configures structured logging (Serilog/.NET, structlog/Python)

Switch between material variants based on conditions or state

Handle packet loss gracefully with retransmission and error correction

Use when converting mathematical graphs, circuit diagrams, phasor diagrams, waveforms, block diagrams, or other technical figures into Obsidian TikZJax-renderable TikZ code blocks for dark-theme notes.

Expert knowledge for AI intelligence collection — OSINT methodology, entity extraction, knowledge graphs, change detection, and sentiment analysis

Usage `rdc:fixit <description>` — sanctioned bypass for quick fixes under 5 files / 30 min. Creates a minimal work item, makes the fix, commits, and closes. The ONLY alternative to rdc:build. Intended for typos, config patches, hotfixes, dependency bumps, small CSS fixes, and similar immediate, low-scope changes.

Deep travel research workflow for museums and ancient architecture. Input a city name and it auto-generates a structured knowledge document (org-mode) plus portable reference cards (PNG). Covers historical background, museum highlights, archaeological significance, and architectural heritage. Use when the user says '旅行研究', '博物馆功课', '古建功课', 'travel research', '出发前功课', or provides a city name with the intent to perform deep cultural travel preparation.

Modern Python development with Python 3.12+, Django, FastAPI, async patterns, and production best practices. Use for Python projects, APIs, data processing, or automation scripts.