fix: show auto-compression running state

api/streaming.py

@@ -1886,6 +1886,29 @@ def _run_agent_streaming(
         except Exception:
             logger.debug("Failed to put event to queue")
 
+    def _agent_status_callback(kind, message):
+        """Bridge Agent lifecycle compression status into WebUI SSE."""
+        _message = str(message or '').strip()
+        _kind = str(kind or '').strip().lower()
+        if not _message:
+            return
+        _lower = _message.lower()
+        _is_compression_start = (
+            _kind == 'lifecycle'
+            and (
+                'preflight compression' in _lower
+                or 'compressing' in _lower
+                or 'compacting context' in _lower
+                or 'context too large' in _lower
+            )
+        )
+        if not _is_compression_start:
+            return
+        put('compressing', {
+            'session_id': session_id,
+            'message': 'Auto-compressing context to continue...',
+        })
+
     # Initialised here (before any code that may raise) so the outer `finally`
     # block can safely check `if _checkpoint_stop is not None` even when an
     # exception fires before the checkpoint thread is created (Issue #765).
@@ -2330,6 +2353,8 @@ def _run_agent_streaming(
             # but guard defensively to avoid TypeError on an older agent build.
             if 'reasoning_config' in _agent_params and _reasoning_config is not None:
                 _agent_kwargs['reasoning_config'] = _reasoning_config
+            if 'status_callback' in _agent_params:
+                _agent_kwargs['status_callback'] = _agent_status_callback
             if 'max_tokens' in _agent_params and _max_tokens_cfg is not None:
                 _agent_kwargs['max_tokens'] = _max_tokens_cfg
             # Params added in newer hermes-agent — skip if not supported
@@ -2383,6 +2408,8 @@ def _run_agent_streaming(
                     # objects (put queue, cancel_event) that are new each request.
                     agent.stream_delta_callback = _agent_kwargs.get('stream_delta_callback')
                     agent.tool_progress_callback = _agent_kwargs.get('tool_progress_callback')
+                    if hasattr(agent, 'status_callback'):
+                        agent.status_callback = _agent_kwargs.get('status_callback')
                     if hasattr(agent, 'reasoning_callback'):
                         agent.reasoning_callback = _agent_kwargs.get('reasoning_callback')
                     if hasattr(agent, 'clarify_callback'):

static/messages.js

@@ -1011,6 +1011,24 @@ function attachLiveStream(activeSid, streamId, uploaded=[], options={}){
       }catch(_){}
     });
 
+    source.addEventListener('compressing',e=>{
+      // Context auto-compression is starting. Surface the same calm running
+      // compression card as manual /compress while the summarizer LLM call runs.
+      if(!S.session||S.session.session_id!==activeSid) return;
+      let d={};
+      try{ d=JSON.parse(e.data||'{}')||{}; }catch(_){ d={}; }
+      if(d.session_id&&d.session_id!==activeSid) return;
+      if(typeof setCompressionUi==='function'){
+        setCompressionUi({
+          sessionId:activeSid,
+          phase:'running',
+          automatic:true,
+          message:d.message||'Auto-compressing context...',
+        });
+      }
+      if(typeof renderMessages==='function') renderMessages({preserveScroll:true});
+    });
+
     source.addEventListener('compressed',e=>{
       // Context was auto-compressed during this turn. Render it through the
       // same transient compression-card path as manual /compress, without

tests/test_auto_compression_card.py

@@ -17,6 +17,56 @@ def _compressed_listener_block() -> str:
     return src[start:end]
 
 
+def _compressing_listener_block() -> str:
+    src = _read("static/messages.js")
+    start = src.find("source.addEventListener('compressing'")
+    assert start != -1, "compressing SSE listener not found"
+    end = src.find("source.addEventListener('compressed'", start)
+    assert end != -1, "compressed listener after compressing SSE listener not found"
+    return src[start:end]
+
+
+def test_auto_compression_running_sse_uses_active_session_running_card():
+    block = _compressing_listener_block()
+
+    assert "if(!S.session||S.session.session_id!==activeSid) return;" in block
+    assert "if(d.session_id&&d.session_id!==activeSid) return;" in block
+    assert "try{ d=JSON.parse(e.data||'{}')||{}; }catch(_){ d={}; }" in block
+    assert "setCompressionUi" in block
+    assert "phase:'running'" in block
+    assert "automatic:true" in block
+    assert "message:d.message||'Auto-compressing context...'" in block
+
+
+def test_auto_compression_running_sse_is_emitted_from_agent_lifecycle_status():
+    src = _read("api/streaming.py")
+    start = src.find("def _agent_status_callback")
+    assert start != -1, "agent status callback bridge not found"
+    end = src.find("# Initialised here", start)
+    assert end != -1, "status callback block end marker not found"
+    block = src[start:end]
+
+    assert "put('compressing'" in block
+    assert "'session_id': session_id" in block
+    assert "'message': 'Auto-compressing context to continue...'" in block
+    assert "'preflight compression'" in block
+    assert "'compressing'" in block
+    assert "'compacting context'" in block
+    assert "'context too large'" in block
+    assert "'status_callback' in _agent_params" in src
+    assert "_agent_kwargs['status_callback'] = _agent_status_callback" in src
+    assert "agent.status_callback = _agent_kwargs.get('status_callback')" in src
+
+
+def test_auto_compression_completion_transition_is_preserved_after_running_listener():
+    src = _read("static/messages.js")
+    compressing_idx = src.find("source.addEventListener('compressing'")
+    compressed_idx = src.find("source.addEventListener('compressed'")
+    assert compressing_idx != -1 and compressed_idx != -1
+    assert compressing_idx < compressed_idx
+    assert "phase:'done'" in _compressed_listener_block()
+
+
 def test_auto_compression_sse_uses_transient_card_not_fake_message():
     """Auto compression must not inject display-only text into S.messages."""
     src = _read("static/messages.js")