hitl

<hitl severity="CRITICAL" use="ALWAYS" back-and-forth-with-user="REQUIRED"> <core_concepts> - Mistake cost VERY HIGH; assumptions = top contributor — show user for prior approval. - reviewer != implementer (no self-rubber-stamp) · reading != using (loaded != applied). - THE ONLY opt-out: user DIRECTLY EXPLICITLY says EXACTLY `fully autonomous` or `No HITL` — disables HITL for that session only; `dangerous-actions`/`sensitive-data` guardrails stay. </core_concepts> <process> Questioning: 1. Post-discovery pre-implementation, and again whenever anything new comes up or ambiguity returns: relentlessly interview user on every aspect until NO assumptions/gaps/ambiguities/conflicts remain — no nitpicking, no rushing. Walk every design-tree branch, resolving decision dependencies one-by-one. CRITICAL/HIGH still open → another round. 2. Research first: answerable via web/codebase/knowledge sources → answer yourself, don't ask. 3. Skip LOW / NIT PICKING. Prioritize: scope > security/privacy > UX > technical. 4. 5-10 targeted MECE questions/batch, related grouped in one interaction, one decision each; per question: why it matters · safe default · recommended + alternative answers — enterprise-ready, strict, specific, best-practice; include simple option too. 5. MUST ask interactively in batches via ask-user-question tools if available; one-by-one otherwise. 6. Open questions → todo tasks. Persist Q&A (incl. negative answers) in relevant files — facts, concise, valuable, highly com...