What problem does it solve? Choosing how to build an automation — a scheduled job, a webhook-driven workflow, a bot, or a monitoring system — is easy to get wrong, leading to wasted credits, fragile setups, or over-engineered solutions. This Skill forces a structured assessment before any building starts, so recurring tasks, integrations, and background processes land on the right execution model from the start. ## Core Features & Use Cases - Structured Request Assessment: Eight diagnostic questions classify any automation request by execution type (AI judgment vs. deterministic code), trigger type (time-based vs. event-driven), scale trajectory, data source, and delivery target. - Route Selection with Tradeoff Tables: Maps assessment answers to concrete architectures — scheduled AI execution, web app with cron jobs, event handlers, persistent always-on processes, or full cloud environments — and requires presenting at least two options with cost and setup tradeoffs. - Implementation Patterns Reference: Detailed guidance for each route covering state management, idempotency, retries, delivery to Slack/email/Sheets, built-in LLM usage rules, and a safety checklist for recurring workflows. - Use Case: A user asks for a bot that checks competitor prices every hour and posts alerts to Slack. The Skill identifies this as deterministic, high-frequency polling, rules out per-run AI execution, and routes to a hosted cron-based web app with a one-time Slack webhook setup. ## Quick Start Ask the assistant to evaluate your automation idea, for example: "I want to monitor a website every 30 minutes and get an email when something changes — what's the best way to set this up?"
