WeKnora/internal/im
wizardchen 1f5970b67e refactor(agent): rebuild multi-turn history from DB, drop llmcontext layer
The Agent ran a parallel context cache (Redis/in-memory) on top of the
messages table to feed multi-turn history into the LLM. That dual-write
caused subtle drift (e.g. compression diverging from DB, system-prompt
swaps lost on restart) and required a separate ClearContext path on
session/IM clear.

Make the messages table the single source of truth:

- Add service.LoadAgentHistory: rebuilds chat.Message history per turn
  from the persisted messages, expanding AgentSteps into proper OpenAI
  assistant_with_tool_calls + tool messages and replaying the canonical
  final answer (with <think> blocks stripped). final_answer tool calls
  are filtered to avoid duplicating the trailing answer.
- Make AgentEngine stateless across turns: drop ContextManager / sessionID
  cache plumbing from the engine, agent service, and CreateAgentEngine
  signature. The engine only uses sessionID for logging/event emission.
- Wire AgentQA to load history from DB on demand using HistoryTurns
  (default 5) when MultiTurnEnabled, otherwise run with empty history.
- Delete the llmcontext package (ContextManager interface, Redis/memory
  storage, factory) and the SessionService.ClearContext API path; IM
  /clear and session message clear no longer need to invalidate cache.

Behavior preserved: KnowledgeQA-mode replay is unchanged (turns with
empty AgentSteps just produce the single canonical assistant message),
and Agent-mode turns now consistently see prior tool calls and results.
2026-05-12 12:26:24 +08:00
..
dingtalk refactor(im): move adapter factories into per-platform subpackages 2026-04-29 20:08:22 +08:00
feishu refactor(im): move adapter factories into per-platform subpackages 2026-04-29 20:08:22 +08:00
mattermost refactor(im): move adapter factories into per-platform subpackages 2026-04-29 20:08:22 +08:00
slack refactor(im): move adapter factories into per-platform subpackages 2026-04-29 20:08:22 +08:00
telegram refactor(im): move adapter factories into per-platform subpackages 2026-04-29 20:08:22 +08:00
wechat fix(im): clamp reconnect backoff to avoid int64 overflow → busy loop 2026-05-05 00:36:18 +08:00
wecom fix(im): clamp reconnect backoff to avoid int64 overflow → busy loop 2026-05-05 00:36:18 +08:00
adapter.go feat(im): add WeChat integration with QR code login and long-polling support 2026-04-13 14:17:57 +08:00
cmd_clear.go feat(im): add slash-command system with /help, /info, /search, /stop, /clear 2026-03-20 15:49:59 +08:00
cmd_help.go feat(im): add slash-command system with /help, /info, /search, /stop, /clear 2026-03-20 15:49:59 +08:00
cmd_info.go feat(im): add slash-command system with /help, /info, /search, /stop, /clear 2026-03-20 15:49:59 +08:00
cmd_search.go feat: Enhance agent tool compatibility and knowledge base filtering 2026-04-23 17:28:00 +08:00
cmd_stop.go feat(im): add slash-command system with /help, /info, /search, /stop, /clear 2026-03-20 15:49:59 +08:00
command.go feat(im): add slash-command system with /help, /info, /search, /stop, /clear 2026-03-20 15:49:59 +08:00
command_registry.go feat(im): add slash-command system with /help, /info, /search, /stop, /clear 2026-03-20 15:49:59 +08:00
content_rewrite_test.go fix(im): rewrite private storage URLs to HTTP in IM channel replies 2026-04-30 11:40:30 +08:00
credentials.go refactor(im): move adapter factories into per-platform subpackages 2026-04-29 20:08:22 +08:00
mode.go refactor(im): move adapter factories into per-platform subpackages 2026-04-29 20:08:22 +08:00
qaqueue.go fix(im): rewrite private storage URLs to HTTP in IM channel replies 2026-04-30 11:40:30 +08:00
ratelimit.go feat(im): add multi-instance support with Redis-based distributed coordination 2026-03-20 15:49:59 +08:00
service.go refactor(agent): rebuild multi-turn history from DB, drop llmcontext layer 2026-05-12 12:26:24 +08:00
service_quote_test.go fix(im): anti-hallucination for non-text quotes and unprocessable media messages 2026-04-03 11:34:22 +08:00
session_test.go [NELO-1102] feat(im): add thread-based session mode for IM channels 2026-03-30 10:43:14 +08:00
session_title_test.go fix(sessions): review fixes for keyword search / pinning / IM titles 2026-04-30 16:23:00 +08:00
stream_test.go feat: enhance IM integration with streaming support and documentation 2026-03-16 02:26:17 +08:00
think.go feat(im): add AI card streaming for DingTalk and improve IM platform robustness 2026-03-24 22:12:08 +08:00
types.go feat(im): add WeChat integration with QR code login and long-polling support 2026-04-13 14:17:57 +08:00
types_test.go [NELO-1102] feat(im): add thread-based session mode for IM channels 2026-03-30 10:43:14 +08:00