发现 Skills

Pytest Test Generator - Auto-activating skill for Test Automation. Triggers on: pytest test generator, pytest test generator Part of the Test Automation skill category.

Diagnose and fix the top Replit errors: container sleep, port binding, Nix failures, DB limits. Use when encountering Replit errors, debugging failed deployments, or troubleshooting workspace and hosting issues. Trigger with phrases like "replit error", "fix replit", "replit not working", "debug replit", "replit broken", "replit deploy failed".

Build a full-stack web app on Replit with Express/Flask, PostgreSQL, Auth, and deployment. Use when creating a new production app on Replit from scratch, building the primary user-facing workflow, or following Replit best practices. Trigger with phrases like "build replit app", "replit full stack", "replit web app", "create replit project", "replit express flask".

Evaluate and improve interface usability using heuristic analysis. Use when the user mentions "usability audit", "UX review", "users are confused", "heuristic evaluation", "form usability", or "navigation problems". Covers Nielsen's 10 heuristics, severity ratings, and information architecture. For visual design fixes, see refactoring-ui. For conversion-focused audits, see cro-methodology. Trigger with 'ux', 'heuristics'.

Guide 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).

Python computations in science and engineering (pycse) - helps with scientific computing tasks including nonlinear regression, uncertainty quantification, design of experiments (DOE), Latin hypercube sampling, surface response modeling, and neural network-based UQ with DPOSE. Use when working with numerical optimization, data fitting, experimental design, or uncertainty analysis.

Systematic code refactoring with risk mitigation. Trigger: When refactoring legacy code, migrating technologies, or resolving technical debt.

Deep research on technical topics using EXA tools with intelligent two-tier caching. Use when user asks to research a topic, investigate best practices, look up information, find patterns, or explore architectures. Also invoked by /research command. Triggers: "research", "look up", "investigate", "deep dive", "find information about", "what are best practices for", "how do others implement".

Use when a proposal has unanimous support or relies on a single high-impact assumption—constructs the strongest possible counter-argument (Steel Man) and runs a Pre-Mortem.

Standardize DeepSeek API direct-model integration for OpenAI-compatible chat, thinking mode, JSON output, streaming, and tool calling. Use when Codex needs to choose a DeepSeek model, wire chat-completions request/response flows, expose thinking controls, JSON object output, function/tool calling, stream controls, model selectors, connection profiles, or shared progress and error states. Do not use for non-DeepSeek providers, Agent/App orchestration, or prompt-only work with no integration change.

Standardize Google Gemini API direct-model integration for chat, vision, Nano Banana image generation and editing, thinking levels, structured outputs, streaming, and function calling. Use when Codex needs to choose a Gemini model, wire generateContent or streamGenerateContent flows, expose thinking controls, JSON schema output, tool/function calling, image input, image generation, connection profiles, model selectors, or shared progress and error states. Do not use for Vertex AI, non-Gemini Google Cloud APIs, non-Gemini providers, Agent/App orchestration, or prompt-only work with no integration change.

Testing patterns for GPUI applications

Manage secrets safely in code and workflows. Use when handling API keys, tokens, passwords, certificates, environment variables, or secret rotation.

Create a new feature module following a Clean Architecture structure. Use when adding a new business domain or feature module to the project.

Google Stitch design platform skill bundle. Covers design system extraction, prompt enhancement, iterative site building, React component conversion, Remotion walkthrough videos, and shadcn/ui component integration. Activate when user mentions Stitch, design-to-code, DESIGN.md, shadcn/ui, or iterative site generation.

Writes or updates the architecture document for a system — system-level technical decisions, component boundaries, tech stack, and data flow patterns. Use when making technology choices, defining component structure, deciding integration approaches, or when architecture decisions emerge during planning conversations.

Build project documentation sites with MkDocs static site generator. Use when a user mentions mkdocs, documentation site, docs site, project documentation, or wants to create, configure, build, or deploy documentation using Markdown. Covers installation, configuration, theming, plugins, and deployment.

Cloud CLI patterns for GCP and AWS. Use when running bq queries, gcloud commands, aws commands, or making decisions about cloud services. Covers BigQuery cost optimization and operational best practices.

Developer reference for GoHighLevel API v2 — OAuth 2.0, private tokens, endpoint reference, rate limits, webhooks, and building marketplace apps.

Narrative RPG game master for collaborative storytelling. Use for tabletop RPG sessions, solo narrative games, or any real-time collaborative fiction requiring scene management, NPC portrayal, and story coherence.

Direct REST API access to UniProt. Protein searches, FASTA retrieval, ID mapping, Swiss-Prot/TrEMBL. For Python workflows with multiple databases, prefer bioservices (unified interface to 40+ services). Use this for direct HTTP/REST work or UniProt-specific control.

Pythonic wrapper around RDKit with a simplified interface and sensible defaults. Preferred for standard drug discovery tasks including SMILES parsing, standardization, descriptor calculation, fingerprints, clustering, 3D conformer generation, and parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters, use RDKit directly.

Systematically evaluate scholarly work using the ScholarEval framework, providing structured assessment across research quality dimensions including problem formulation, methodology, analysis, and writing with quantitative scoring and actionable feedback.

Hugging Face Transformers for loading Hub models, running pipeline inference, text generation, and Trainer fine-tuning on NLP, vision, audio, and multimodal tasks. Use when working with AutoModel, pipelines, tokenizers, or TrainingArguments—not for general ML outside the Transformers library.