← ClaudeAtlas

office-hourslisted

Force assumptions into the open before code is written. 6 mentor-style forcing questions sharpen requirements + reframe imprecise goals. Goal-type mode-mapping (Revenue / Infra / Content / Personal / Side-quest) adjusts question depth. Complements /scope by adding an explicit Q&A loop on the user's framing itself.
0xUrsanomics/utopia-os · ★ 2 · AI & Automation · score 79
Install: claude install-skill 0xUrsanomics/utopia-os
# Office Hours. Forcing-Question Layer Before Build ## Why this exists Requirements written by the user are usually imprecise, hypothetical, or carry hidden assumptions. The fix is not better implementation. it's 60 seconds of forcing questions before any code is written. This skill makes those questions explicit and repeatable. Complement to `/scope` (which surfaces assumptions in the agent's plan). `office-hours` interrogates the USER's framing itself. Use BEFORE `/scope` when the request feels mushy. ## When to fire - User describes a vague product idea ("I want to build X for Y users") - Pre-implementation when the brief reads like marketing language ("seamless", "better", "modern") - User says "I'm thinking about..." or "should I build..." - Before any new project intake, new client/project scoping Skip when the ask is concrete, scoped, and has a measurable definition of done. Skip when an explicit "just do it" override is given. ## Pre-reading (consult once before Step 1) Read §1 (the structured-user-question decision-brief format) of `knowledge/gstack-cross-cutting-patterns.md`. The 6 questions are walked one at a time as decision briefs, not freestyle. ## Procedure ### Step 1. Context gathering (quiet) Read in this order. Don't print findings. ground yourself. - `CLAUDE.md` if it exists - `memory/USER.md` + the active persona memory file - `memory/SOUL.md` for voice register if output-shaped - Recent git log: `git log --oneline -15` - Glob/grep for related