Aska 8603839711 feat(ai-core): trace instrumentation, LLM thinker, decay scheduler ...

- orchestrator: Tracer interface with spans for intent analysis, persona,
  subsession dispatch, context build, LLM streaming and each tool call;
  traces finish when the stream completes — the admin Trace waterfall
  finally receives data
- background: Thinker generates real LLM thoughts (post-chat + silence
  triggers, recent history as context, empty/overlong outputs discarded);
  thoughts broadcast to web clients as 'thinking' WS messages; pipeline
  postChatHook triggers post-chat thinking after each round
- memory: ApplyDecayAll + ListUserIDs with a 6h decay/purge cron in main
  (was implemented but never scheduled)

2026-08-20 20:57:27 +08:00

..

adapter.go

feat(gateway): wire real services into Gateway API handlers

2026-08-12 12:04:31 +08:00

bridge_llm.go

feat(memory): add LLM-driven memory extraction from conversations

2026-08-12 18:43:04 +08:00

embedder.go

fix(memory): rewrite dedup with n-gram keyword overlap + embedder

2026-08-12 18:24:04 +08:00

extractor_test.go

feat(memory): add LLM-driven memory extraction from conversations

2026-08-12 18:43:04 +08:00

extractor.go

feat(memory): add LLM-driven memory extraction from conversations

2026-08-12 18:43:04 +08:00

model.go

feat(memory): hybrid vector search via embedder

2026-08-20 20:57:13 +08:00

repository.go

feat(ai-core): trace instrumentation, LLM thinker, decay scheduler

2026-08-20 20:57:27 +08:00

service_test.go

fix(memory): rewrite dedup with n-gram keyword overlap + embedder

2026-08-12 18:24:04 +08:00

service.go

feat(ai-core): trace instrumentation, LLM thinker, decay scheduler

2026-08-20 20:57:27 +08:00