For architectural changes, design memo before code, even in auto-mode
Operating an agent in continuous-execution mode where the user has authorised a substantial feature ('ship X') but the feature involves architectural choices with open tradeoffs
User said 'ship #166' — a multi-attribution / many-to-many data-model change with explicitly-open tradeoffs in the issue body. I interpreted 'ship' as a directive to execute and started adding columns + writing migrations. The user interrupted with 'wait can you clarify how this PR works?' before I'd gotten further than the schema. Wrote out the design memo, surfaced two real open questions (primary-attribution behaviour for groups, direction display on user's own page), and stopped for confirmation. Realised the mistake: the directive was fine on small fixes earlier in the session, but for a structural change with named open tradeoffs in the source ticket, jumping straight to code skips the most important step — confirming the architectural choices being baked in. The cost of writing a design memo first is ~5 minutes; the cost of building the wrong shape and rebuilding is hours.
When the user says 'ship X' but X is structural — new schema, new contract between layers, new invariant — write the design memo first and pause, regardless of how clear the directive sounded. Specifically: enumerate the worked examples (what gets attributed where), surface every tradeoff the source ticket left unresolved, name the read-path changes downstream consumers will notice. Cost of clarifying upfront is bounded; cost of building the wrong shape compounds. Auto-mode is a license to skip planning on routine work, not on architectural commits.