fix(chat): three concurrency holes found reviewing my own patch (v4.6.77)

1. chat.py renamed the temp file AFTER releasing the flock, so two writers could each
   read the pre-replace snapshot and the later rename silently dropped the earlier
   message. os.replace now happens inside the lock.
2. The agent cursor was global. After hermes-reset it pointed past the end of the new
   empty thread, so every read returned nothing — the agent would think it had been
   ignored. Key is now per thread and reset clears every chat cursor.
3. chat-relay.sh had no overlap guard: a slow agent run (up to 5 min) could collide with
   the next tick and answer the same thread twice. flock -n on /tmp/chat-relay.lock.

Verified: stale cursor 'webx11_chat_agent_cursor_oldthread00000' set to 9999999999999,
hermes-reset -> zero cursor keys remain, read returns {messages:[], note:'no new messages...'}.

6 files changed, 21 insertions, 9 deletions

diff --git a/main.html b/main.html
index 23ab761..8cc0848 100644
--- a/main.html
+++ b/main.html
@@ -1349,13 +1349,13 @@
     </script>
 
     <!-- v4.5.63: A2A scratchpad fixes + agent-register arg fix -->
-<script src="modules/core/core-bundle.js?v=v4.6.76"></script>
+<script src="modules/core/core-bundle.js?v=v4.6.77"></script>
 
     <!-- Async dependency-ordered module loader now only initializes terminal and loads optional ext modules -->
-<script src="modules/core/module-loader.js?v=v4.6.76"></script>
+<script src="modules/core/module-loader.js?v=v4.6.77"></script>
 
     <!-- Critical ext bundle loads AFTER webx11 exists so registerModule calls work -->
-<script src="modules/ext/critical-ext-bundle.js?v=v4.6.76"></script>
+<script src="modules/ext/critical-ext-bundle.js?v=v4.6.77"></script>
 
 <script>
 // ═══ ONE TRUE VERSION SOURCE ═══
diff --git a/modules/core/core-bundle.js b/modules/core/core-bundle.js
index 697b5ec..90e2593 100644
--- a/modules/core/core-bundle.js
+++ b/modules/core/core-bundle.js
@@ -11098,8 +11098,9 @@ buildSystemPrompt() {
                                 // runs every 30 s and advances the shared cursor, so a read that used it
                                 // returned nothing — the answer was printed for the human and the model
                                 // never saw it. Two readers, two cursors. First read is from 0.
+                                const acKey = 'webx11_chat_agent_cursor_' + (st.thread || 'new');
                                 let ac = 0;
-                                try { ac = parseInt(localStorage.getItem('webx11_chat_agent_cursor') || '0', 10) || 0; } catch (e) { ac = 0; }
+                                try { ac = parseInt(localStorage.getItem(acKey) || '0', 10) || 0; } catch (e) { ac = 0; }
                                 const resp = await fetch(
                                     '/spaces/newos4/api/chat/poll.php' + (ac ? '?since=' + ac : ''),
                                     { headers: { 'X-Chat-Token': st.token } });
@@ -11107,7 +11108,7 @@ buildSystemPrompt() {
                                 if (!resp.ok || !j || !j.success) {
                                     return JSON.stringify({ error: (j && j.error) || ('HTTP ' + resp.status) });
                                 }
-                                if (j.cursor) { try { localStorage.setItem('webx11_chat_agent_cursor', String(j.cursor)); } catch (e) {} }
+                                if (j.cursor) { try { localStorage.setItem(acKey, String(j.cursor)); } catch (e) {} }
                                 const msgs = (j.messages || []).slice(-10).map(m => ({
                                     author: m.author, ts: m.ts, text: m.text
                                 }));
@@ -20789,6 +20790,10 @@ class DebugSupervisorModule {
         try {
             localStorage.removeItem('webx11_chat_token');
             localStorage.removeItem('webx11_chat_state');
+            // a stale agent cursor would point past the end of the new, empty thread
+            Object.keys(localStorage)
+                .filter(k => k.indexOf('webx11_chat_agent_cursor') === 0)
+                .forEach(k => localStorage.removeItem(k));
             this.terminal.print('🔄 Chat token rotated — a new thread starts with your next message', 'success');
         } catch (e) {
             this.terminal.print('❌ ' + e.message, 'error');
diff --git a/modules/core/module-aiassist.js b/modules/core/module-aiassist.js
index cefff5d..c3ebe1f 100644
--- a/modules/core/module-aiassist.js
+++ b/modules/core/module-aiassist.js
@@ -2711,8 +2711,9 @@ buildSystemPrompt() {
                                 // runs every 30 s and advances the shared cursor, so a read that used it
                                 // returned nothing — the answer was printed for the human and the model
                                 // never saw it. Two readers, two cursors. First read is from 0.
+                                const acKey = 'webx11_chat_agent_cursor_' + (st.thread || 'new');
                                 let ac = 0;
-                                try { ac = parseInt(localStorage.getItem('webx11_chat_agent_cursor') || '0', 10) || 0; } catch (e) { ac = 0; }
+                                try { ac = parseInt(localStorage.getItem(acKey) || '0', 10) || 0; } catch (e) { ac = 0; }
                                 const resp = await fetch(
                                     '/spaces/newos4/api/chat/poll.php' + (ac ? '?since=' + ac : ''),
                                     { headers: { 'X-Chat-Token': st.token } });
@@ -2720,7 +2721,7 @@ buildSystemPrompt() {
                                 if (!resp.ok || !j || !j.success) {
                                     return JSON.stringify({ error: (j && j.error) || ('HTTP ' + resp.status) });
                                 }
-                                if (j.cursor) { try { localStorage.setItem('webx11_chat_agent_cursor', String(j.cursor)); } catch (e) {} }
+                                if (j.cursor) { try { localStorage.setItem(acKey, String(j.cursor)); } catch (e) {} }
                                 const msgs = (j.messages || []).slice(-10).map(m => ({
                                     author: m.author, ts: m.ts, text: m.text
                                 }));
diff --git a/modules/core/module-debug-supervisor.js b/modules/core/module-debug-supervisor.js
index 6768df1..79c9acc 100644
--- a/modules/core/module-debug-supervisor.js
+++ b/modules/core/module-debug-supervisor.js
@@ -173,6 +173,10 @@ class DebugSupervisorModule {
         try {
             localStorage.removeItem('webx11_chat_token');
             localStorage.removeItem('webx11_chat_state');
+            // a stale agent cursor would point past the end of the new, empty thread
+            Object.keys(localStorage)
+                .filter(k => k.indexOf('webx11_chat_agent_cursor') === 0)
+                .forEach(k => localStorage.removeItem(k));
             this.terminal.print('🔄 Chat token rotated — a new thread starts with your next message', 'success');
         } catch (e) {
             this.terminal.print('❌ ' + e.message, 'error');
diff --git a/modules/core/module-loader.js b/modules/core/module-loader.js
index e52a846..b177626 100644
--- a/modules/core/module-loader.js
+++ b/modules/core/module-loader.js
@@ -4,7 +4,7 @@
     // Core module bundle loaded synchronously in main.html.
     // This loader only handles version cache-busting, splash, terminal init, and optional ext modules.
 
-    const LATEST_VER = 'v4.6.76';
+    const LATEST_VER = 'v4.6.77';
     const CACHE = '?v=' + LATEST_VER;
     const stored = localStorage.getItem('webx11_cache_ver');
     if (stored !== LATEST_VER) {
diff --git a/scripts/chat.py b/scripts/chat.py
index e19fb6b..5b44b64 100755
--- a/scripts/chat.py
+++ b/scripts/chat.py
@@ -78,9 +78,11 @@ def reply(thread, text):
             with open(tmp, 'w') as fh:
                 json.dump(msgs, fh, ensure_ascii=False)
             os.chmod(tmp, 0o664)
+            # replace INSIDE the lock: doing it after unlock lets a second writer read
+            # the pre-replace snapshot and rename over this message (lost update).
+            os.replace(tmp, path)
         finally:
             fcntl.flock(lock, fcntl.LOCK_UN)
-    os.replace(tmp, path)          # atomic: a poll never sees a half-written file
     return msg