From 1011c950bef934d32eca90095b7f3b5867ec9851 Mon Sep 17 00:00:00 2001
From: ekko <152005280+EKKOLearnAI@users.noreply.github.com>
Date: Wed, 6 May 2026 14:07:13 +0800
Subject: [PATCH] fix: improve chat message display and compression logic
 (#476)

Frontend:
- Filter out assistant messages with empty content in mapHermesMessages()
- Prevents displaying 194/585 empty messages in long sessions
- Reduces visual clutter and improves UX

Backend:
- Add message count limit (<= 200) to compression threshold checks
- Prevents compressing sessions with too many messages
- Improves compression performance and reliability

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 packages/client/src/stores/hermes/chat.ts            | 12 ++++++++++--
 .../server/src/services/hermes/chat-run-socket.ts    |  4 ++--
 2 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/packages/client/src/stores/hermes/chat.ts b/packages/client/src/stores/hermes/chat.ts
index 89d9ee6..3b8005e 100644
--- a/packages/client/src/stores/hermes/chat.ts
+++ b/packages/client/src/stores/hermes/chat.ts
@@ -119,10 +119,18 @@ async function buildContentBlocks(
 }
 
 function mapHermesMessages(msgs: HermesMessage[]): Message[] {
+  // Filter out assistant messages with empty content
+  const filteredMsgs = msgs.filter(m => {
+    if (m.role === 'assistant') {
+      return m.content && m.content.trim() !== ''
+    }
+    return true
+  })
+
   // Build lookups from assistant messages with tool_calls
   const toolNameMap = new Map<string, string>()
   const toolArgsMap = new Map<string, string>()
-  for (const msg of msgs) {
+  for (const msg of filteredMsgs) {
     if (msg.role === 'assistant' && msg.tool_calls) {
       for (const tc of msg.tool_calls) {
         if (tc.id) {
@@ -134,7 +142,7 @@ function mapHermesMessages(msgs: HermesMessage[]): Message[] {
   }
 
   const result: Message[] = []
-  for (const msg of msgs) {
+  for (const msg of filteredMsgs) {
     // Skip assistant messages that only contain tool_calls (no meaningful content)
     if (msg.role === 'assistant' && msg.tool_calls?.length && !msg.content?.trim()) {
       // Emit a tool.started message for each tool call
diff --git a/packages/server/src/services/hermes/chat-run-socket.ts b/packages/server/src/services/hermes/chat-run-socket.ts
index c160160..a8e8d25 100644
--- a/packages/server/src/services/hermes/chat-run-socket.ts
+++ b/packages/server/src/services/hermes/chat-run-socket.ts
@@ -578,7 +578,7 @@ export class ChatRunSocket {
               logger.info('[context-compress] session=%s: snapshot at %d, %d new messages, assembled ~%d tokens (threshold %d)',
                 session_id, snapshot.lastMessageIndex, newMessages.length, totalTokens, triggerTokens)
               // triggerTokens
-              if (totalTokens <= triggerTokens) {
+              if (totalTokens <= triggerTokens && newMessages.length <= 200) {
                 // Under threshold — use assembled context directly, no LLM call needed
                 history = [
                   { role: 'user', content: SUMMARY_PREFIX + '\n\n' + snapshot.summary },
@@ -684,7 +684,7 @@ export class ChatRunSocket {
             } else if (history.length > 4) {
               // No snapshot — check if raw history exceeds threshold
 
-              if (totalTokens <= triggerTokens) {
+              if (totalTokens <= triggerTokens && history.length <= 200) {
                 // Under threshold — use raw history as-is
                 logger.info('[context-compress] session=%s: %d messages, ~%d tokens — under threshold, skip', session_id, history.length, totalTokens)
               } else {