plannotator-setup-goal
Setup Goal
Turn an idea into a goal package at goals/<slug>/ through structured discovery, user interview, and codebase exploration.
Phases
1. Rearticulate
State back what the user wants in your own words. If the conversation already has rich context, summarize it. If the goal is bare or vague, do minimal shallow exploration of the codebase to ground your understanding. Keep it to 2-3 sentences. Wait for the user to confirm or correct before continuing.
Create goals/<slug>/ once the slug is clear. Keep working JSON files and final documents there. JSON preserves provenance and iteration state; markdown is the human-readable authoritative goal package.
Browser session patience: After launching an interview or facts command, wait until the user submits, dismisses, or asks you to stop. Do not close, kill, restart, refresh, or open a second session because the UI is idle. If a rerun is needed, wait for the previous session to end, update its working JSON file, and launch from that file.
For a vague goal with interdependent decisions, suggest an optional one-at-a-time grilling pass. Run it when requested, with a recommended answer for each question. Fold its resolved decisions into the bundle below; skip the bundle if grilling fully resolves scope.
2. Interview Bundle
Build a compact bundle of questions that can derive every outcome fact. Infer answers already clear from the request, conversation, or codebase. Only ask where the user's judgment is needed. Prefer fewer, higher-leverage questions over exhaustive confirmation.