hailin db8617dda8 refactor(agents): remove per-intent hard truncation, keep 2000-char safety net ... Hard-coded INTENT_MAX_ANSWER_LENGTH limits caused mid-sentence truncation and content loss. Length control now relies on prompt + schema description + LLM-Judge (3 layers). Only a 2000-char safety net remains for extreme edge cases. Also simplified followUp: non-question followUp is now always appended (prevents model content split from silently dropping text). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>		2026-02-07 10:32:53 -08:00
..
agents	refactor(agents): remove per-intent hard truncation, keep 2000-char safety net	2026-02-07 10:32:53 -08:00
claude	fix(tools): 修复 coordinator-tools 与 immigration-tools 之间的 input_schema 不一致	2026-02-07 01:31:38 -08:00
database/postgres	feat(agents): add admin-configurable evaluation gate for agent loop quality control	2026-02-06 18:56:52 -08:00
knowledge	feat(agents): persist evaluation gate failures as system experiences for cross-conversation learning	2026-02-06 21:28:56 -08:00
payment	feat(payment): P2 — 订单管理增强，支持取消订单和订单详情查询	2026-02-07 01:23:28 -08:00