productivity▌

Diagnose whether your AI workflows are context stuffing or context engineering, then implement structured practices to improve output quality. \n \n Distinguishes between context stuffing (volume-based) and context engineering (structure-based), with five diagnostic questions to identify Context Hoarding Disorder \n Guides two-layer memory architecture: short-term conversational memory plus long-term persistent memory via vector databases for semantic retrieval \n Implements the Research→Plan→Re

Structured 30-60-90 day onboarding framework drawing from 14 product leaders' approaches. \n \n Covers listening tours, milestone planning, early wins, and belonging through pairing rather than isolation on day one \n Emphasizes defining success at 90 days, 1 year, and 2 years before onboarding begins, plus identifying golden rituals to teach by first Friday \n Includes relationship design conversations, documented management philosophy, and early ownership assignment to accelerate ramp-up to 30

Manage local Airflow development environment with Astro CLI commands. \n \n Start, stop, restart, and kill local Airflow containers; default credentials are admin/admin with webserver at http://localhost:8080 \n View logs for all components or specific services (scheduler, webserver) with real-time follow option \n Access container shells and run Airflow CLI commands directly via astro dev bash and astro dev run \n Troubleshoot common issues including port conflicts, startup failures, package er

Use this skill when the user:

Headless data tables with server-side pagination, filtering, sorting, and virtualization for Cloudflare Workers + D1. \n \n Supports three rendering modes: client-side with core row model, server-side with manual state management, and virtualized rendering for 1000+ row datasets \n Includes column/row pinning, row expanding with nested data, and row grouping with built-in aggregation functions (sum, min, max, mean, etc.) \n Integrates with TanStack Query for coordinated data fetching and state s

Systematic checklist for reviewing code across functionality, security, performance, and maintainability. \n \n Covers six review areas: context understanding, functionality verification, code quality assessment, security checks, performance analysis, and test coverage validation \n Includes concrete examples of common issues (SQL injection, missing validation, unclear naming) paired with corrected versions \n Provides pre-review, per-section, and git-specific checklists to ensure nothing is mis

Structured product requirements documents with problem statements, user stories, and success metrics. \n \n Guides PRD structure across eight sections: problem statement, goals, non-goals, user stories, requirements (P0/P1/P2), success metrics, open questions, and timeline considerations \n Provides frameworks for user story writing, MoSCoW requirement prioritization, and acceptance criteria in Given/When/Then format \n Includes guidance on defining leading and lagging success metrics with speci

Self-contained Convex packages with isolated databases, functions, and type-safe exports for reuse across projects. \n \n Includes isolated database tables, queries, mutations, actions, and TypeScript types within a single component package \n Supports multiple component instances in a single app with custom naming and configuration options \n Provides optional React hooks for type-safe frontend integration and simplified consumer APIs \n Publishes to npm with proper peer dependencies, TypeScrip

Publish articles to Toutiao with persistent login and browser-based publishing. \n \n One-time QR code login with automatic session persistence; subsequent publishes use saved credentials without re-authentication \n Interactive publishing mode opens authenticated browser directly to the publish page, or fully automated mode via command-line arguments (title, content file, cover image) \n Automatic title optimization to meet Toutiao's 2–30 character requirement; truncates or pads as needed \n Se

Generate podcast episodes with 1-2 AI speakers discussing a topic. Supports quick overviews, deep analysis, and debate formats. Input can be a topic description, URL(s), or text. Output is a full audio episode with transcript.