Giorgio Gilestro 0254515989 docs: refine LLM-CSV spec — keep real sample row, drop user attribution ... - Drop first_seen_user_id; sample is anonymous by construction - Rename sample_dummy → sample_row, store the upload's first real data row verbatim (one row, no totals, no other positions, no link to a user). Narrow, deliberate exception to the "no holdings persisted" invariant — gives the operator material for hand-writing future native parsers. - Drop the cache self-heal behaviour; operator owns eviction. Reinforce the non-goal of auto-promoting learned formats to code.		2026-05-27 11:21:43 +02:00
..
plans	docs: implementation plan for beta + paid-gap rollout	2026-05-25 18:28:39 +02:00
specs	docs: refine LLM-CSV spec — keep real sample row, drop user attribution	2026-05-27 11:21:43 +02:00