发现 Skills

Create custom Cascade agent configurations for specialized tasks. Activate when users mention "custom cascade agent", "specialized ai agent", "domain-specific cascade", "agent configuration", or "custom ai behavior". Handles custom agent creation and configuration. Use when working with windsurf cascade agents functionality. Trigger with phrases like "windsurf cascade agents", "windsurf agents", "windsurf".

Access AlphaFold's 200M+ AI-predicted protein structures. Retrieve structures by UniProt ID, download PDB/mmCIF files, analyze confidence metrics (pLDDT, PAE), for drug discovery and structural biology.

Reference the Kubernetes API. Useful for understanding what is possible via the Kubernetes REST API.

When starting a new PR task: update main, clean up local branches, and create a feature branch.

Use when giving or receiving performance feedback, conducting 1:1s, or managing developmental conversations.

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

Design abstract strategy games with perfect information, no randomness, and strategic depth. Use when designing a board game, exploring abstract strategy games, brainstorming game mechanics, or evaluating game balance. Keywords: board game, game design, strategy, mechanics, balance.

Diagnose flat dialogue, same-voice characters, and lack of subtext. Use when conversations feel wooden, characters sound alike, or dialogue only does one thing at a time.

Use for systematic debugging and root cause analysis with Codex. Triggers on "codex debug", "codex root cause", "have codex investigate this bug", "codex find the issue".

Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.

GPU-accelerate Python code using CuPy, Numba CUDA, Warp, cuDF, cuML, cuGraph, KvikIO, cuCIM, cuxfilter, cuVS, cuSpatial, and RAFT. Use whenever the user mentions GPU/CUDA/NVIDIA acceleration, or wants to speed up NumPy, pandas, scikit-learn, scikit-image, NetworkX, GeoPandas, or Faiss workloads. Covers physics simulation, differentiable rendering, mesh ray casting, particle systems (DEM/SPH/fluids), vector/similarity search, GPUDirect Storage file IO, interactive dashboards, geospatial analysis, medical imaging, and sparse eigensolvers. Also use when you see CPU-bound Python code (loops, large arrays, ML pipelines, graph analytics, image processing) that would benefit from GPU acceleration, even if not explicitly requested.

Standard single-cell RNA-seq analysis pipeline. Use for QC, normalization, dimensionality reduction (PCA/UMAP/t-SNE), clustering, differential expression, and visualization. Best for exploratory scRNA-seq analysis with established workflows. For deep learning models use scvi-tools; for data format questions use anndata.

Upgrades subagent dispatch to Agent Teams for do-plan, do-execute, and do-discuss phases. Use when running do-plan, do-execute, or do-discuss at standard/deep depth.

Manages ServiceNow update sets. Use for creating, retrieving, previewing, committing. Triggers - customizations, deployments.

Semantic SEO framework for reducing Cost of Retrieval (CoR). Use when creating SEO content, topical maps, content briefs, semantic content networks, knowledge graphs, website audits, or optimizing for search engines. Triggers on SEO strategy, content optimization, E-A-T, entity optimization, search visibility, ranking improvement, content architecture, and semantic search.

When the user wants to choose an SEO-friendly domain for a new site—brand vs keyword domain, TLD selection, or domain best practices. Also use when the user mentions "domain choice," "pick domain," "SEO-friendly domain," "brand domain," "EMD," "PMD," "exact match domain," "partial match domain," "TLD," ".ai domain," ".com vs .io," "domain length," "domain history," or "defensive domain registration." For migration, use rebranding-strategy.

Write actually funny comedy: stand-up bits, monologues, one-liners, observational stories, roasts, and comic essays. Use when the user asks for jokes, bits, stand-up material, roast jokes, funny stories, comedy writing in any form, or anything requiring genuine humor. Also use when asked to be funny, make me laugh, or write something with comedic intent.

Invoke this skill when building any retrieval-augmented generation (RAG) system. Covers document loaders, RecursiveCharacterTextSplitter, embeddings (OpenAI), and vector stores (Chroma, FAISS, Pinecone).

Feishu instant messaging: sending and receiving messages and managing group chats. Send and reply to messages, search chat history, manage group members, upload and download images and files, and manage emoji reactions. Use when a user needs to send a message, view or search chat history, download files from a chat, or view group members.

Use this skill when a user requests drawing charts in Lark cloud documents or wants to use the Lark Whiteboard to draw architecture diagrams, flowcharts, mind maps, sequence diagrams, or other visualizations.

Core coding principles and design aphorisms for writing maintainable code. Use when writing, reviewing, or refactoring code. Use it anytime you need to write code.

Upgrades existing websites and apps to premium quality. Audits current design, identifies generic AI patterns, and applies high-end design standards without breaking functionality. Works with any CSS framework or vanilla CSS.

Skill with multiple supporting files

Research how a feature works in reference libraries