automation-advisor

What it does

Guides you through a structured 8-phase decision framework to evaluate whether a task is worth automating. Scores four dimensions (frequency, time, error cost, longevity), applies seven override checks for high-stakes scenarios, calculates break-even ROI, and generates an Obsidian-formatted analysis report. Available as a Claude Code skill, console app, or voice-enabled web interface.

Key features

• 4-dimension scoring matrix — frequency × time × error cost × longevity produces a quantified automation score with clear decision thresholds (automate now / maybe / stay manual)
• 7 override checks — catches edge cases where high scores still shouldn’t be automated: regulated industries, creative work, rapidly changing requirements, learning fundamentals
• Validation patterns — for high-stakes automation, recommends human-in-the-loop, confidence thresholds, audit trails, or staged rollout approaches
• Break-even calculator — estimates build time investment, weekly time saved, and weeks to ROI
• Voice interface — Groq Whisper STT input and browser Web Speech API output for hands-free evaluation
• Web UI — dark-theme responsive interface with real-time progress tracking and multi-user sessions

When to use

When deciding whether to automate a repetitive task — transforms “this feels tedious” into a quantified recommendation with clear next steps and risk assessment.