productivity▌

CRITICAL — 开始前 MUST 先用 Read 工具读取 ../lark-shared/SKILL.md，其中包含认证、权限处理

Deploy and manage full-stack web apps with Firebase App Hosting using Next.js, Angular, and other supported frameworks. \n \n Requires Firebase project on Blaze pricing plan; supports Server-Side Rendering (SSR) and Incremental Static Regeneration (ISR) workflows \n Deploy via firebase.json configuration with optional apphosting.yaml for backend setup, or enable automated \"git push to deploy\" through GitHub integration \n Includes secret management via CLI commands for secure access to sensiti

Web search optimized for AI agents with real-time data retrieval. \n \n Single command interface ( scripts/search ) for querying the web with results tailored for agent consumption \n Supports JSON output formatting for structured data integration into agent workflows \n Requires Tavily API key for authentication; free API key available from tavily.com \n

Routing hub for Convex backend development skills covering functions, schemas, realtime, storage, agents, and security. \n \n Routes to 11 specialized sub-skills via commands like /convex-functions , /convex-realtime , /convex-agents , and /convex-migrations \n Covers core patterns: queries, mutations, actions, reactive subscriptions, webhooks, file storage, and scheduled tasks \n Includes security-focused skills for audits and best-practice checklists \n Recommended entry point is /convex-best-

Unified AI SDK for Dart enabling code generation, structured outputs, tools, flows, and agents. \n \n Provides core APIs for generation, tool definition, flow orchestration, embeddings, and streaming with a single interface \n Includes 8+ plugins for LLM providers (Google Gemini, Anthropic Claude, OpenAI GPT), Firebase AI, Model Context Protocol, Chrome browser integration, and HTTP server hosting via Shelf \n Built-in CLI with local development UI for flow execution, tracing, model experimentat

Interprets shorthand, pseudo-code, and natural language descriptions to generate production-quality code. \n \n Assesses collaborator technical expertise (high/medium/low confidence) to determine how much interpretation and correction is needed \n Processes special shorthand notation marked with start-shorthand and end-shorthand boundaries; always removes ()=> lines when implementing \n Applies expert judgment to translate incomplete or imperfect descriptions into robust code, handling typos, te

Export Google Sheets spreadsheets as CSV files for backup or local processing. \n \n Requires gws-sheets and gws-drive skills as prerequisites \n Offers three export approaches: retrieve spreadsheet metadata, export via Drive API with MIME type specification, or read sheet values directly in CSV format \n Supports flexible range selection and sheet targeting for granular control over exported data \n

Sanitize user prompts through Google Model Armor safety templates. \n \n Requires a Model Armor template resource name and accepts text input via flag, stdin, or full JSON request body \n Designed for inbound prompt safety; use the companion +sanitize-response command for outbound response filtering \n Integrates with Google Cloud authentication and global flags defined in gws-shared \n

Prerequisite setup for Firebase agent integration: Node.js, CLI authentication, and MCP server installation. \n \n Verifies Node.js v20+ installation and guides setup via nvm or official installers for macOS, Linux, and Windows \n Confirms Firebase CLI availability through npx firebase-tools and authenticates via firebase login with remote environment support \n Directs agent-specific MCP server and skill installation for Gemini CLI, Antigravity, Claude Code, Cursor, GitHub Copilot, and other pl

Socratic mentoring for junior developers through guided questions, never direct answers. \n \n Guides learners through structured questioning rather than solving problems, building autonomy and reasoning skills across fundamentals, debugging, testing, security, and architecture \n Includes the PEAR Loop (Plan, Explore, Analyze, Rewrite) for Copilot-assisted learning and teaches prompt engineering with the CTEX formula for better AI interactions \n Provides progressive clue systems from light gui