feat(agent): enhance streaming diagnostics and event tracking

- Added diagnostics for streaming events in the AgentEngine, including chunk counts and response type distributions.
- Implemented logging for emitted event types during the thinking process to aid in debugging and performance analysis.
- Updated the session handler to provide more informative warnings when fallback answers are emitted due to non-streaming behavior.
- Introduced a dedicated HTTP client for Ollama to improve connection handling and timeout management during streaming calls.
This commit is contained in:
wizardchen
2026-03-24 16:26:23 +08:00
committed by lyingbug
parent 1c9503b063
commit 2b56c4fbac
5 changed files with 78 additions and 12 deletions
@@ -450,7 +450,11 @@ func (h *AgentStreamHandler) handleComplete(ctx context.Context, evt event.Event
// This guards against edge cases where the LLM stops without calling final_answer.
if h.finalAnswer == "" && data.FinalAnswer != "" {
logger.GetLogger(h.ctx).Warnf(
"No answer events were streamed, emitting fallback answer (len=%d)", len(data.FinalAnswer),
"No answer events were streamed, emitting fallback answer (len=%d). "+
"This typically happens when: (1) model stopped naturally and content was sent as thought events, "+
"or (2) Ollama model returned tool calls non-incrementally. "+
"total_steps=%d, total_duration_ms=%d",
len(data.FinalAnswer), data.TotalSteps, data.TotalDurationMs,
)
fallbackID := fmt.Sprintf("answer-fallback-%d", time.Now().UnixMilli())
if err := h.streamManager.AppendEvent(h.ctx, h.sessionID, h.assistantMessageID, interfaces.StreamEvent{