diff --git a/pkg/agent/loop.go b/pkg/agent/loop.go
index 5558f7c0e..693f2227b 100644
--- a/pkg/agent/loop.go
+++ b/pkg/agent/loop.go
@@ -759,13 +759,7 @@ func (al *AgentLoop) maybeSummarize(agent *AgentInstance, sessionKey, channel, c
 		if _, loading := al.summarizing.LoadOrStore(summarizeKey, true); !loading {
 			go func() {
 				defer al.summarizing.Delete(summarizeKey)
-				if !constants.IsInternalChannel(channel) {
-					al.bus.PublishOutbound(bus.OutboundMessage{
-						Channel: channel,
-						ChatID:  chatID,
-						Content: "Memory threshold reached. Optimizing conversation history...",
-					})
-				}
+				logger.Debug("Memory threshold reached. Optimizing conversation history...")
 				al.summarizeSession(agent, sessionKey)
 			}()
 		}
diff --git a/pkg/providers/openai_compat/provider.go b/pkg/providers/openai_compat/provider.go
index a8d244d4a..087d3506e 100644
--- a/pkg/providers/openai_compat/provider.go
+++ b/pkg/providers/openai_compat/provider.go
@@ -115,8 +115,12 @@ func (p *Provider) Chat(
 	// with the same key and reuse prefix KV cache across calls.
 	// The key is typically the agent ID — stable per agent, shared across requests.
 	// See: https://platform.openai.com/docs/guides/prompt-caching
+	// Prompt caching is only supported by OpenAI-native endpoints.
+	// Gemini and other providers reject unknown fields, so skip for non-OpenAI APIs.
 	if cacheKey, ok := options["prompt_cache_key"].(string); ok && cacheKey != "" {
-		requestBody["prompt_cache_key"] = cacheKey
+		if !strings.Contains(p.apiBase, "generativelanguage.googleapis.com") {
+			requestBody["prompt_cache_key"] = cacheKey
+		}
 	}
 
 	jsonData, err := json.Marshal(requestBody)