发现 Skills

Send and read Google Chat messages to Spaces and DMs. Resolve person names via ClickHouse. Use when user says 'send gchat', 'read google chat', 'DM [person]', 'gchat [space]', 'отправь в чат', 'прочитай чат'.

Build Marketing Mix Models with PyMC, adstock and saturation transforms, and mandatory out-of-sample validation. Use when user says "build MMM model", "построй MMM модель", "marketing mix model", or asks to train MMM on client data.

Troubleshoot Restate Cloud failures via Admin API, including stuck invocations, determinism errors, and scheduler recovery. Use when user says "restate troubleshooting", "fix restate invocation", "почини restate", or asks to diagnose Restate errors.

Detects XSS, CSRF, prototype pollution and other security issues; supports React/Vue/Angular

Agent Build Machine Module Architecture Guide (Go Language), covering the Agent startup process, heartbeat mechanism, task acquisition and execution, upgrades and updates, and interaction with Dispatch. It is used when users develop Agent features, modify heartbeat logic, handle task execution, or implement Agent upgrades.

BK-CI Database Design Specifications and Table Structure Guide, covering naming conventions, field type selection, index design, sharding strategies, and data archiving. Used when users design database tables, optimize indexes, plan sharding strategies, or perform database architecture design.

CloudBase Run backend development rules (Function mode / Container mode). Use this skill when deploying backend services that require long-lived connections, multi-language support, custom environments, or AI agent development.

Use this skill when developing Node.js backend services or CloudBase cloud functions (Express/Koa/NestJS, serverless, backend APIs) that need AI capabilities. Features text generation (generateText), streaming (streamText), AND image generation (generateImage) via @cloudbase/node-sdk ≥3.16.0. Built-in models include Hunyuan (hunyuan-2.0-instruct-20251111 recommended), DeepSeek (deepseek-v3.2 recommended), and hunyuan-image for images. This is the ONLY SDK that supports image generation. NOT for browser/Web apps (use ai-model-web) or WeChat Mini Program (use ai-model-wechat).

Standard software engineering workflow for requirement analysis, technical design, and task planning. Use this skill when developing new features, complex architecture designs, multi-module integrations, or projects involving database/UI design.

Professional UI design and frontend interface guidelines. Use this skill when creating web pages, mini-program interfaces, prototypes, or any frontend UI components that require distinctive, production-grade design with exceptional aesthetic quality.

Orchestrate Tenzir workflows. Use when executing Tenzir workflows.

Apache Airflow is a platform for programmatically authoring, scheduling, and monitoring workflows. Learn to write DAGs, use operators, set up connections, configure scheduling, and deploy with Docker Compose.

Configure Caddy as a web server and reverse proxy — automatic HTTPS, reverse proxy, load balancing, file server, redirects, headers, rate limiting, and API configuration. Use when tasks involve serving websites, proxying to backend services, automatic TLS certificate management, or replacing Nginx with a simpler configuration.

Build production APIs with FastAPI advanced patterns. Use when a user asks to implement dependency injection, background tasks, WebSockets, middleware, database sessions, or structured error handling in FastAPI.

Write well-structured git commit messages following the Conventional Commits standard. Use when a user asks to commit changes, write a commit message, stage and commit, improve a commit message, or follow commit conventions. Analyzes diffs to produce accurate, meaningful commit messages.

Build high-performance RPC services with gRPC and Protocol Buffers. Use when a user asks to create gRPC services, define protobuf schemas, implement streaming RPCs, build microservice communication, set up service-to-service calls, implement bidirectional streaming, add interceptors/middleware to gRPC, generate client stubs, handle gRPC errors, implement health checks, configure load balancing, or build gRPC-Web for browser clients. Covers unary, server/client/bidirectional streaming, interceptors, deadlines, metadata, reflection, and production patterns.

Extract text and structured data from images using Vision AI (OCR). Use when: reading text from screenshots, extracting data from scanned documents, converting images of tables/forms/charts to structured text.

When the user needs to generate, validate, refresh, or debug JSON Web Tokens. Use when the user mentions "JWT," "access token," "refresh token," "token rotation," "token expiration," "token validation," "bearer token," or "decode JWT." Handles secure token lifecycle including signing, verification, refresh rotation, and revocation. For full auth system design, see auth-system-setup.

Configure and use Kibana for Elasticsearch data visualization, dashboard creation, and log exploration. Use when a user needs to build dashboards with Lens, write KQL queries, set up data views, create visualizations, or configure Kibana spaces and role-based access.

Build Model Context Protocol (MCP) servers that connect AI agents to external services and data sources. Use when a user asks to create an MCP server, build an MCP tool, connect an AI agent to an API, create a tool server for Claude, build MCP resources, or expose a database/service via MCP. Generates TypeScript or Python MCP servers with tools, resources, and prompts following the official MCP specification.

Self-host, configure, scale, and manage n8n instances. Use when a user asks to install n8n with Docker, set up n8n in production, configure n8n environment variables, scale n8n with workers, back up n8n workflows, restore n8n data, manage n8n via CLI or REST API, monitor n8n, troubleshoot n8n issues, set up n8n with PostgreSQL, or deploy n8n with a reverse proxy. For designing n8n workflows and nodes, see n8n-workflow.

Assists with finding security vulnerabilities in web applications using OWASP ZAP. Use when configuring automated scans, writing scan policies, integrating security scanning into CI/CD pipelines, or analyzing results for OWASP Top 10 vulnerabilities like XSS, SQL injection, and CSRF. Trigger words: owasp zap, security scan, vulnerability scanner, penetration testing, zap-baseline, active scan, passive scan.

When the user needs to design database schemas and queries optimized for real-time applications. Use when the user mentions "chat database," "message storage," "real-time sync," "message history," "unread count," "cursor pagination," "event sourcing," or "live data." Handles schema design for messaging, activity feeds, notifications, and collaborative apps with efficient pagination and sync. For WebSocket transport, see websocket-builder.

Design and build RESTful APIs. Use when a user asks to create a REST API, design API endpoints, implement CRUD operations, add authentication to an API, handle pagination and filtering, set up API versioning, generate OpenAPI/Swagger docs, implement rate limiting, build middleware, handle file uploads, add CORS, implement HATEOAS, or structure an Express/Fastify/Koa/Flask/FastAPI backend. Covers HTTP methods, status codes, resource modeling, error handling, validation, and production deployment patterns.