Merge pull request #1724 from nesquena/stage-303

v0.51.6 — 5-PR full-sweep batch
2026-05-25 19:20:16 +00:00 · 2026-05-05 15:11:06 -07:00
parent 0ea3dfbdd1 23bca0d955
commit a6e2bbb263
17 changed files with 466 additions and 22 deletions
@@ -1,5 +1,41 @@
 # Hermes Web UI -- Changelog

+## [v0.51.6] — 2026-05-05 — 5-PR full-sweep batch
+
+### Added
+
+- **PR #1719** by @Michaelyklam — Show active elapsed time in compact activity (closes #1716). Adds an in-progress elapsed counter while the agent is still working, complementing the already-shipped post-completion duration. Backend `/api/chat/start` now returns `pending_started_at` timestamp; UI uses that as the durable source of truth (instead of a browser-local timer that resets on rerender/reconnect). The compact Activity-row timer settles back to the existing post-completion duration display when the turn finishes. Cleanup timer paths attached to `setBusy(false)`, `clearLiveToolCards()`, `removeThinking()` so the counter stops on every terminal path (turn ends, session switch, error).
+
+### Fixed
+
+- **PR #1717** by @ai-ag2026 — Preserve imported session lineage visibility. Three independent fixes for the CLI/messaging session import path: (a) preserve `parent_session_id` when importing CLI/messaging sessions into WebUI sidecars (lineage was being dropped); (b) avoid shrinking sidebar `message_count` when CLI metadata has fewer messages than a repaired/aggregate sidecar (the sidebar was reverting to the shorter count); (c) prefer the longer WebUI sidecar transcript for messaging `/api/session` responses when it contains recovered visible history. 4 new regression tests cover lineage import, read-only imports, sidebar counts, and the recovered-sidecar transcript-selection path.
+- **PR #1718** by @Michaelyklam — Preserve Activity count across chat focus changes (closes #1715). Root cause: `loadSession()` restored `S.toolCalls` from the per-session `INFLIGHT` cache, then replayed those tools through `appendLiveToolCard()` BEFORE restoring `S.activeStreamId`. `appendLiveToolCard()` intentionally no-ops without `S.activeStreamId`, so the replayed tools were dropped from the compact Activity group after focus changed. Fix: restore `S.activeStreamId` BEFORE the tool replay loop. Source-level regression assertion pins the new ordering.
+- **PR #1720** by @Michaelyklam — Fix backend tool snippet cap for "Show more" (closes #1714). Frontend already had logic to preview long tool snippets at ~800 chars and reveal the rest with "Show more", but the backend was truncating persisted tool snippets to 200 chars — so the frontend threshold could never be reached. Raises the persisted snippet cap from 200 → 4000 chars (conservative; medium tool outputs can use the existing affordance, huge outputs are still bounded so session JSON doesn't balloon). Per-issue maintainer-confirmed direction.
+- **PR #1722** by @ai-ag2026 — Suppress stale preserved task lists. After context compaction or reload, the UI was re-rendering the most recent preserved compression task-list card from history even after the actual todo state had moved on (all items completed/cancelled). Stale tasks reappeared as if still pending. Fix: only treat `pending` and `in_progress` todos as "active" when deciding whether to keep the preserved task list visible. Regression test covers the stale-preserved-task-list suppression path. Handles the `latestTodos === null` fallback correctly (no fresh todo tool message found → keep showing the preserved card, original behavior).
+
+### Tests
+
+4527 → **4537 passing** (+10 regression tests across the 5 PRs). 0 regressions. Full suite ~149s.
+
+### Pre-release verification
+
+- Stage-303: 5 PRs merged with zero conflicts (each rebased clean against current master). Zero stage-applied edits.
+- All JS files syntax-clean (`node -c static/{messages,sessions,ui}.js`).
+- All Python files syntax-clean (py_compile on every changed file).
+- Live browser walkthrough on port 8789:
+  - PR #1718 ordering fix: `S.activeStreamId` is set BEFORE `appendLiveToolCard()` replay (CORRECT-ORDER verified in source).
+  - PR #1719 `pending_started_at` flows through to messages/UI; elapsed timer code present.
+  - PR #1722 todo state filter present in source.
+  - PR #1717 sidebar module helpers present.
+  - Sidebar scroll holds at 200 (carry-over fix from v0.51.2 preserved).
+  - System health card from v0.51.5 still working in Insights (CPU 15%, RAM 48.3%, disk 33.9%).
+- Opus advisor: SHIP, 6/6 verification clean, 0 MUST-FIX, 0 SHOULD-FIX. Two non-blocking observations:
+  - #1717 "longer sidecar wins" heuristic won't honor explicit CLI-side message deletions (low likelihood for messaging sessions; documented).
+  - #1719 elapsed timer is client-clock-relative; gross browser clock drift will distort live counter (cosmetic; follow-up could send server-clock anchor).
+
+Closes #1714, #1715, #1716.
+
+
 ## [v0.51.5] — 2026-05-05 — 4-PR full-sweep batch

 ### Added
@@ -2,7 +2,7 @@

 > Web companion to the Hermes Agent CLI. Same workflows, browser-native.
 >
-> Last updated: v0.51.5 (May 5, 2026) — 4527 tests collected
+> Last updated: v0.51.6 (May 5, 2026) — 4537 tests collected
 > Test source: `pytest tests/ --collect-only -q`
 > Per-version detail: see [CHANGELOG.md](./CHANGELOG.md)

@@ -1835,8 +1835,8 @@ Bridged CLI sessions:

 ---

-*Last updated: v0.51.5, May 5, 2026*
-*Total automated tests collected: 4527*
+*Last updated: v0.51.6, May 5, 2026*
+*Total automated tests collected: 4537*
 *Regression gate: tests/test_regressions.py*
 *Run: pytest tests/ -v --timeout=60*
 *Source: <repo>/*
@@ -1229,9 +1229,13 @@ def import_cli_session(
    profile=None,
    created_at=None,
    updated_at=None,
+    parent_session_id=None,
 ):
-    """Create a new WebUI session populated with CLI messages.
-    Returns the Session object.
+    """Create a new WebUI session populated with CLI/agent messages.
+
+    Preserve parent_session_id from state.db so imported continuation segments
+    keep their lineage in the WebUI store and sidebar instead of reappearing as
+    detached orphan chats.
    """
    s = Session(
        session_id=session_id,
@@ -1242,6 +1246,7 @@ def import_cli_session(
        profile=profile,
        created_at=created_at,
        updated_at=updated_at,
+        parent_session_id=parent_session_id,
    )
    s.save(touch_updated_at=False)
    return s
@@ -1270,9 +1270,15 @@ def _merge_cli_sidebar_metadata(ui_session: dict, cli_meta: dict) -> dict:
    if cli_meta.get("last_message_at") is not None:
        merged["last_message_at"] = cli_meta["last_message_at"]
    if cli_meta.get("message_count") is not None:
-        merged["message_count"] = cli_meta["message_count"]
+        merged["message_count"] = max(
+            _numeric_count(merged.get("message_count")),
+            _numeric_count(cli_meta.get("message_count")),
+        )
    elif cli_meta.get("actual_message_count") is not None:
-        merged["message_count"] = cli_meta["actual_message_count"]
+        merged["message_count"] = max(
+            _numeric_count(merged.get("message_count")),
+            _numeric_count(cli_meta.get("actual_message_count")),
+        )

    if cli_meta.get("title"):
        current_title = merged.get("title")
@@ -2622,7 +2628,13 @@ def handle_get(handler, parsed) -> bool:
            _t3 = _time.monotonic()
            if load_messages:
                if is_messaging_session and cli_messages:
-                    _all_msgs = cli_messages
+                    sidecar_messages = getattr(s, "messages", []) or []
+                    # Recovery/aggregate sidecars can intentionally contain a
+                    # longer visible conversation than the single state.db
+                    # segment for this messaging session id. Prefer the longer
+                    # sidecar so repaired WebUI history is not hidden behind the
+                    # canonical per-segment transcript.
+                    _all_msgs = sidecar_messages if len(sidecar_messages) > len(cli_messages) else cli_messages
                else:
                    _all_msgs = s.messages
            else:
@@ -5986,7 +5998,11 @@ def _handle_chat_start(handler, body):
        daemon=True,
    )
    thr.start()
-    response = {"stream_id": stream_id, "session_id": s.session_id}
+    response = {
+        "stream_id": stream_id,
+        "session_id": s.session_id,
+        "pending_started_at": s.pending_started_at,
+    }
    if normalized_model:
        response["effective_model"] = model
    if model_provider:
@@ -7661,6 +7677,7 @@ def _handle_session_import_cli(handler, body):
                "raw_source": existing.raw_source or cli_meta.get("raw_source") or cli_meta.get("source_tag"),
                "session_source": existing.session_source or cli_meta.get("session_source"),
                "source_label": existing.source_label or cli_meta.get("source_label"),
+                "parent_session_id": existing.parent_session_id or cli_meta.get("parent_session_id"),
            }
            for attr, value in updates.items():
                if getattr(existing, attr, None) != value:
@@ -7702,6 +7719,7 @@ def _handle_session_import_cli(handler, body):
    cli_thread_id = None
    cli_session_key = None
    cli_platform = None
+    cli_parent_session_id = None
    cli_read_only = False
    for cs in get_cli_sessions():
        if cs["session_id"] == sid:
@@ -7720,6 +7738,7 @@ def _handle_session_import_cli(handler, body):
            cli_thread_id = cs.get("thread_id")
            cli_session_key = cs.get("session_key")
            cli_platform = cs.get("platform")
+            cli_parent_session_id = cs.get("parent_session_id")
            cli_read_only = bool(cs.get("read_only"))
            break

@@ -7750,6 +7769,7 @@ def _handle_session_import_cli(handler, body):
            "raw_source": cli_raw_source or cli_source_tag,
            "session_source": cli_session_source,
            "source_label": cli_source_label,
+            "parent_session_id": cli_parent_session_id,
            "read_only": True,
            "messages": msgs,
            "tool_calls": [],
@@ -7764,6 +7784,7 @@ def _handle_session_import_cli(handler, body):
        profile=profile,
        created_at=created_at,
        updated_at=updated_at,
+        parent_session_id=cli_parent_session_id,
    )
    if cron_project_id:
        s.project_id = cron_project_id
@@ -1379,16 +1379,22 @@ def _merge_display_messages_after_agent_result(previous_display, previous_contex
    return merged


-def _tool_result_snippet(raw) -> str:
-    """Extract a compact result preview from a stored tool message payload."""
+_TOOL_RESULT_SNIPPET_MAX = 4000
+
+
+def _tool_result_snippet(raw, limit: int = _TOOL_RESULT_SNIPPET_MAX) -> str:
+    """Extract a bounded result preview from a stored tool message payload."""
+    if limit <= 0:
+        return ''
    text = str(raw or '')
    try:
-        data = json.loads(text)
+        data = raw if isinstance(raw, dict) else json.loads(text)
        if isinstance(data, dict):
-            return str(data.get('output') or data.get('result') or data.get('error') or text)[:200]
+            preview = data.get('output') or data.get('result') or data.get('error') or text
+            text = str(preview)
    except Exception:
        pass
-    return text[:200]
+    return text[:limit]


 def _truncate_tool_args(args, limit: int = 6) -> dict:
@@ -243,6 +243,9 @@ async function send(){
    }
    streamId=startData.stream_id;
    S.activeStreamId = streamId;
+    if(S.session&&typeof startData.pending_started_at==='number'){
+      S.session.pending_started_at=startData.pending_started_at;
+    }
    if(S.session&&S.session.session_id===activeSid){
      S.session.active_stream_id = streamId;
    }
@@ -430,6 +430,10 @@ async function loadSession(sid){
    S.messages=INFLIGHT[sid].messages;
    S.toolCalls=(INFLIGHT[sid].toolCalls||[]);
    S.busy=true;
+    // appendLiveToolCard() is guarded by S.activeStreamId; restore it before
+    // replaying persisted live tools so the compact Activity count survives
+    // switching away from and back to an active chat (#1715).
+    S.activeStreamId=activeStreamId;
    syncTopbar();renderMessages();appendThinking();loadDir('.');
    clearLiveToolCards();
    if(typeof placeLiveToolCardsHost==='function') placeLiveToolCardsHost();
@@ -440,7 +444,6 @@ async function loadSession(sid){
    startApprovalPolling(sid);
    if(typeof startClarifyPolling==='function') startClarifyPolling(sid);
    if(typeof _fetchYoloState==='function') _fetchYoloState(sid);
-    S.activeStreamId=activeStreamId;
    if(INFLIGHT[sid].reattach&&activeStreamId&&typeof attachLiveStream==='function'){
      INFLIGHT[sid].reattach=false;
      if (_loadingSessionId !== sid) return;
@@ -1420,6 +1420,72 @@ function _formatTurnDuration(seconds){
  if(h)return`${h}h ${m}m`;
  return`${m}m ${s}s`;
 }
+function _formatActiveElapsedTimer(seconds){
+  const n=Number(seconds);
+  if(!Number.isFinite(n)||n<0)return'';
+  const total=Math.max(0,Math.floor(n));
+  const m=Math.floor(total/60);
+  const s=total%60;
+  return`${String(m).padStart(2,'0')}:${String(s).padStart(2,'0')}`;
+}
+let _activityElapsedTimer=null;
+let _activityElapsedTimerGroup=null;
+function _activityElapsedStartedAt(group){
+  if(!group)return null;
+  const raw=(group.dataset&&group.dataset.turnStartedAt!==undefined&&group.dataset.turnStartedAt!=='')
+    ?group.dataset.turnStartedAt
+    :(S.session&&S.session.pending_started_at);
+  const started=Number(raw);
+  return Number.isFinite(started)&&started>0?started:null;
+}
+function _activityElapsedLabel(group){
+  const started=_activityElapsedStartedAt(group);
+  if(!started)return'';
+  return _formatActiveElapsedTimer((Date.now()/1000)-started);
+}
+function _setActivityElapsedStartedAt(group){
+  if(!group||group.getAttribute('data-live-tool-call-group')!=='1')return;
+  const started=_activityElapsedStartedAt(group);
+  if(started)group.setAttribute('data-turn-started-at',String(started));
+}
+function _updateActiveActivityElapsedTimer(){
+  const group=_activityElapsedTimerGroup;
+  if(!group||!group.isConnected||group.getAttribute('data-live-tool-call-group')!=='1'){
+    _clearActivityElapsedTimer();
+    return;
+  }
+  const durationEl=group.querySelector('.tool-call-group-duration');
+  const label=_activityElapsedLabel(group);
+  if(label){
+    group.setAttribute('data-active-turn-elapsed',label);
+  }else{
+    group.removeAttribute('data-active-turn-elapsed');
+  }
+  if(durationEl){
+    durationEl.textContent=label?`Working ${label}`:'';
+    durationEl.style.display=label?'':'none';
+  }
+}
+function _startActivityElapsedTimer(group){
+  if(!group||group.getAttribute('data-live-tool-call-group')!=='1')return;
+  _setActivityElapsedStartedAt(group);
+  if(_activityElapsedTimerGroup&&_activityElapsedTimerGroup!==group)_clearActivityElapsedTimer();
+  _activityElapsedTimerGroup=group;
+  _updateActiveActivityElapsedTimer();
+  if(!_activityElapsedTimer)_activityElapsedTimer=setInterval(_updateActiveActivityElapsedTimer,1000);
+}
+function _clearActivityElapsedTimer(){
+  if(_activityElapsedTimer){
+    clearInterval(_activityElapsedTimer);
+    _activityElapsedTimer=null;
+  }
+  if(_activityElapsedTimerGroup&&_activityElapsedTimerGroup.isConnected){
+    _activityElapsedTimerGroup.removeAttribute('data-active-turn-elapsed');
+    const durationEl=_activityElapsedTimerGroup.querySelector('.tool-call-group-duration');
+    if(durationEl){durationEl.textContent='';durationEl.style.display='none';}
+  }
+  _activityElapsedTimerGroup=null;
+}

 const _MOBILE_CONFIG_BASE_LABEL='Workspace, model, reasoning, and context settings';

@@ -2438,6 +2504,7 @@ function setBusy(v){
  S.busy=v;
  updateSendBtn();
  if(!v){
+    if(typeof _clearActivityElapsedTimer==='function') _clearActivityElapsedTimer();
    setStatus('');
    setComposerStatus('');
    const sid=_queueDrainSid||(S.session&&S.session.session_id);
@@ -3727,7 +3794,9 @@ function ensureActivityGroup(inner, opts){
    if(anchor&&anchor.parentElement===inner) anchor.insertAdjacentElement('afterend', group);
    else inner.appendChild(group);
  }
+  if(live) _setActivityElapsedStartedAt(group);
  _syncToolCallGroupSummary(group);
+  if(live) _startActivityElapsedTimer(group);
  return group;
 }
 function _compressionStateForCurrentSession(){
@@ -3966,9 +4035,29 @@ function _preservedCompressionTaskListCardHtml(m, open=false){
 function _preservedCompressionTaskListCardsHtml(messages){
  return (messages||[]).map(m=>_preservedCompressionTaskListCardHtml(m, false)).join('');
 }
+function _latestTodoToolItems(messages){
+  for(let i=(messages||[]).length-1;i>=0;i--){
+    const m=messages[i];
+    if(!m||m.role!=='tool') continue;
+    try{
+      const payload=typeof m.content==='string'?JSON.parse(m.content):m.content;
+      if(payload&&Array.isArray(payload.todos)) return payload.todos;
+    }catch(_){ }
+  }
+  return null;
+}
+function _hasActiveTodoItems(items){
+  return Array.isArray(items) && items.some(item=>{
+    const status=String(item&&item.status||'').trim().toLowerCase();
+    return status==='pending'||status==='in_progress';
+  });
+}
 function _latestPreservedCompressionTaskListMessages(messages){
  const latest=[...(messages||[])].reverse().find(m=>_isPreservedCompressionTaskListMessage(m));
-  return latest?[latest]:[];
+  if(!latest) return [];
+  const latestTodos=_latestTodoToolItems(messages);
+  if(Array.isArray(latestTodos) && !_hasActiveTodoItems(latestTodos)) return [];
+  return [latest];
 }
 function _isSameLocalDay(dateA, dateB){
  return dateA.getFullYear()===dateB.getFullYear()
@@ -4807,9 +4896,17 @@ function _syncToolCallGroupSummary(group){
  }
  if(list) list.textContent=parts.join(' · ')||'tools / thinking';
  if(durationEl){
-    const durationText=_formatTurnDuration(group.dataset.turnDuration);
-    durationEl.textContent=durationText?`Done in ${durationText}`:'';
-    durationEl.style.display=durationText?'':'none';
+    if(group.getAttribute('data-live-tool-call-group')==='1'){
+      const activeText=_activityElapsedLabel(group);
+      if(activeText) group.setAttribute('data-active-turn-elapsed',activeText);
+      else group.removeAttribute('data-active-turn-elapsed');
+      durationEl.textContent=activeText?`Working ${activeText}`:'';
+      durationEl.style.display=activeText?'':'none';
+    }else{
+      const durationText=_formatTurnDuration(group.dataset.turnDuration);
+      durationEl.textContent=durationText?`Done in ${durationText}`:'';
+      durationEl.style.display=durationText?'':'none';
+    }
  }
  if(badge) badge.textContent=String(total);
 }
@@ -4896,6 +4993,7 @@ function appendLiveToolCard(tc){
 }

 function clearLiveToolCards(){
+  if(typeof _clearActivityElapsedTimer==='function') _clearActivityElapsedTimer();
  const inner=_assistantTurnBlocks($('liveAssistantTurn'));
  if(inner) inner.querySelectorAll('.tool-call-group[data-live-tool-call-group],.tool-card-row[data-live-tid]').forEach(el=>el.remove());
  // Reset the per-turn user expand intent so the next turn starts at the
@@ -5699,7 +5797,10 @@ function removeThinking(){
  if(blocks) blocks.querySelectorAll('.agent-activity-thinking').forEach(el=>el.remove());
  if(blocks) blocks.querySelectorAll('.tool-call-group[data-agent-activity-group="1"]').forEach(group=>{
    _syncToolCallGroupSummary(group);
-    if(!group.querySelector('.tool-card-row,.agent-activity-thinking')) group.remove();
+    if(!group.querySelector('.tool-card-row,.agent-activity-thinking')){
+      if(typeof _clearActivityElapsedTimer==='function') _clearActivityElapsedTimer();
+      group.remove();
+    }
  });
  if(turn&&blocks&&!blocks.children.length) turn.remove();
 }
@@ -155,6 +155,20 @@ def test_preserved_task_list_attaches_once_per_render():
    assert "(!preservedCompressionTaskCardsAttached&&(!referenceMessage||compressionState)&&preservedCompressionTaskMessages.length)" in src


+def test_preserved_task_list_is_suppressed_when_latest_todo_state_has_no_active_items():
+    src = _read("static/ui.js")
+    start = src.find("function _latestTodoToolItems")
+    assert start != -1, "latest todo state helper not found"
+    end = src.find("function _isSameLocalDay", start)
+    assert end != -1, "preserved-task-list helper block end not found"
+    helpers = src[start:end]
+
+    assert "if(payload&&Array.isArray(payload.todos)) return payload.todos;" in helpers
+    assert "function _hasActiveTodoItems" in helpers
+    assert "status==='pending'||status==='in_progress'" in helpers
+    assert "if(Array.isArray(latestTodos) && !_hasActiveTodoItems(latestTodos)) return [];" in helpers
+
+
 def test_preserved_task_list_rendering_does_not_mutate_history():
    src = _read("static/ui.js")
    start = src.find("function _isPreservedCompressionTaskListMessage")
@@ -0,0 +1,34 @@
+import json
+
+
+def test_import_cli_session_preserves_parent_session_id():
+    from api.models import import_cli_session, SESSION_DIR, Session
+
+    parent_id = 'parent_lineage_001'
+    child_id = 'child_lineage_001'
+
+    # Ensure clean fixture state for direct model-level import.
+    for sid in (parent_id, child_id):
+        try:
+            (SESSION_DIR / f'{sid}.json').unlink(missing_ok=True)
+        except Exception:
+            pass
+
+    session = import_cli_session(
+        child_id,
+        'Child Session',
+        [{'role': 'user', 'content': 'hello', 'timestamp': 1.0}],
+        model='test-model',
+        parent_session_id=parent_id,
+        created_at=1.0,
+        updated_at=2.0,
+    )
+
+    assert session.parent_session_id == parent_id
+
+    payload = json.loads((SESSION_DIR / f'{child_id}.json').read_text(encoding='utf-8'))
+    assert payload['parent_session_id'] == parent_id
+
+    loaded = Session.load(child_id)
+    assert loaded.parent_session_id == parent_id
+    assert loaded.compact()['parent_session_id'] == parent_id
@@ -415,7 +415,7 @@ def test_loadSession_inflight_restores_live_tool_cards(cleanup_test_sessions):
    # INFLIGHT branch must call appendLiveToolCard
    inflight_idx = src.find("if(INFLIGHT[sid]){")
    assert inflight_idx >= 0, "INFLIGHT branch not found in loadSession"
-    inflight_block = src[inflight_idx:inflight_idx+500]
+    inflight_block = src[inflight_idx:inflight_idx+900]
    assert "appendLiveToolCard" in inflight_block,         "loadSession INFLIGHT branch must restore live tool cards via appendLiveToolCard"
    assert "clearLiveToolCards" in inflight_block,         "loadSession INFLIGHT branch must clear old live cards before restoring"

@@ -601,6 +601,29 @@ def test_loadSession_inflight_sets_busy_before_renderMessages(cleanup_test_sessi
        "loadSession must set S.busy=true before renderMessages() to avoid duplicate tool cards"


+def test_loadSession_inflight_sets_active_stream_before_replaying_live_tool_cards(cleanup_test_sessions):
+    """#1715: returning to an active chat must replay persisted tool cards.
+
+    appendLiveToolCard() intentionally no-ops unless S.activeStreamId is already
+    set for the viewed streaming session. If loadSession() restores S.toolCalls
+    and replays them before assigning S.activeStreamId, the compact Activity
+    counter drops the previously-seen tools after a focus change.
+    """
+    src = (REPO_ROOT / "static/sessions.js").read_text()
+    inflight_idx = src.find("if(INFLIGHT[sid]){")
+    assert inflight_idx >= 0, "INFLIGHT branch not found in loadSession"
+    inflight_block = src[inflight_idx:inflight_idx+1000]
+    active_pos = inflight_block.find("S.activeStreamId=activeStreamId;")
+    replay_pos = inflight_block.find("appendLiveToolCard(tc);")
+    attach_pos = inflight_block.find("attachLiveStream(sid, activeStreamId")
+    assert active_pos >= 0, "loadSession INFLIGHT branch must restore S.activeStreamId"
+    assert replay_pos >= 0, "loadSession INFLIGHT branch must replay persisted live tool cards"
+    assert active_pos < replay_pos, \
+        "S.activeStreamId must be restored before appendLiveToolCard() replays persisted tools"
+    assert attach_pos < 0 or active_pos < attach_pos, \
+        "S.activeStreamId should also be restored before SSE reattach can deliver more tool events"
+
+
 def test_streaming_bridge_accepts_current_tool_progress_callback_signature(cleanup_test_sessions):
    """R17: api/streaming.py must accept the current Hermes agent callback contract.
    The agent now calls tool_progress_callback(event_type, name, preview, args, **kwargs).
@@ -90,6 +90,7 @@ def test_session_import_cli_refresh_matches_messages_despite_timestamp_type_diff
            self.raw_source = "weixin"
            self.session_source = "messaging"
            self.source_label = "WeChat"
+            self.parent_session_id = None

        def compact(self):
            return {"session_id": session_id, "title": "Imported"}
@@ -141,6 +142,7 @@ def test_session_import_cli_refresh_rejects_prefix_if_non_timing_content_diverge
            self.session_source = "messaging"
            self.source_label = "Telegram"
            self.is_cli_session = True
+            self.parent_session_id = None

        def compact(self):
            return {"session_id": session_id, "title": "Imported"}
@@ -169,3 +171,113 @@ def test_session_import_cli_refresh_rejects_prefix_if_non_timing_content_diverge
    assert response["session"]["messages"] == existing.messages
    assert existing.messages[0]["content"] == "old-prefix"
    assert save_calls == []
+
+
+def test_session_import_cli_preserves_parent_metadata_on_existing_import(monkeypatch):
+    """Refreshing an already-imported CLI session must persist lineage metadata."""
+    import api.routes as routes
+
+    session_id = "existing_parent_lineage_001"
+    parent_id = "root_parent_lineage_001"
+
+    class FakeSession:
+        def __init__(self):
+            self.messages = [{"role": "user", "content": "hello", "timestamp": 1.0}]
+            self.source_tag = "telegram"
+            self.raw_source = "telegram"
+            self.session_source = "messaging"
+            self.source_label = "Telegram"
+            self.parent_session_id = None
+            self.is_cli_session = True
+
+        def compact(self):
+            return {"session_id": session_id, "title": "Imported", "parent_session_id": self.parent_session_id}
+
+        def save(self, touch_updated_at=False):
+            save_calls.append(touch_updated_at)
+
+    save_calls = []
+    existing = FakeSession()
+
+    monkeypatch.setattr(routes.Session, "load", classmethod(lambda _cls, sid: existing if sid == session_id else None))
+    monkeypatch.setattr(routes, "require", lambda body, *keys: None)
+    monkeypatch.setattr(routes, "j", lambda _handler, payload, status=200, extra_headers=None: payload)
+    monkeypatch.setattr(routes, "get_cli_session_messages", lambda sid: existing.messages if sid == session_id else [])
+    monkeypatch.setattr(
+        routes,
+        "get_cli_sessions",
+        lambda: [{
+            "session_id": session_id,
+            "source_tag": "telegram",
+            "raw_source": "telegram",
+            "session_source": "messaging",
+            "source_label": "Telegram",
+            "parent_session_id": parent_id,
+        }],
+    )
+
+    response = routes._handle_session_import_cli(object(), {"session_id": session_id})
+
+    assert response["imported"] is False
+    assert existing.parent_session_id == parent_id
+    assert response["session"]["parent_session_id"] == parent_id
+    assert save_calls == [False]
+
+
+def test_read_only_import_payload_includes_parent_session_id(monkeypatch):
+    """Read-only CLI/session imports should also expose lineage in the payload."""
+    import api.routes as routes
+
+    session_id = "readonly_parent_lineage_001"
+    parent_id = "readonly_root_lineage_001"
+    messages = [{"role": "user", "content": "hello", "timestamp": 1.0}]
+
+    monkeypatch.setattr(routes.Session, "load", classmethod(lambda _cls, sid: None))
+    monkeypatch.setattr(routes, "require", lambda body, *keys: None)
+    monkeypatch.setattr(routes, "bad", lambda _handler, msg, status=400: {"ok": False, "error": msg, "status": status})
+    monkeypatch.setattr(routes, "j", lambda _handler, payload, status=200, extra_headers=None: payload)
+    monkeypatch.setattr(routes, "get_cli_session_messages", lambda sid: messages if sid == session_id else [])
+    monkeypatch.setattr(
+        routes,
+        "get_cli_sessions",
+        lambda: [{
+            "session_id": session_id,
+            "title": "Read-only child",
+            "model": "test-model",
+            "created_at": 1.0,
+            "updated_at": 2.0,
+            "source_tag": "discord",
+            "raw_source": "discord",
+            "session_source": "messaging",
+            "source_label": "Discord",
+            "parent_session_id": parent_id,
+            "read_only": True,
+        }],
+    )
+
+    response = routes._handle_session_import_cli(object(), {"session_id": session_id})
+
+    assert response["imported"] is False
+    assert response["session"]["parent_session_id"] == parent_id
+    assert response["session"]["messages"] == messages
+
+
+def test_merge_cli_sidebar_metadata_keeps_larger_sidecar_message_count():
+    """Sidebar metadata merge should not shrink repaired aggregate sidecar counts."""
+    import api.routes as routes
+
+    merged = routes._merge_cli_sidebar_metadata(
+        {"session_id": "sid", "message_count": 535, "title": "Recovered"},
+        {"session_id": "sid", "message_count": 407, "source_tag": "discord"},
+    )
+
+    assert merged["message_count"] == 535
+
+
+def test_messaging_session_loader_prefers_longer_sidecar_transcript():
+    """Pin the /api/session invariant that repaired sidecars can be longer than state.db segments."""
+    handler = _extract_handler("handle_get")
+    old = "if is_messaging_session and cli_messages:\n                    _all_msgs = cli_messages"
+    assert old not in handler
+    assert "sidecar_messages = getattr(s, \"messages\", []) or []" in handler
+    assert "len(sidecar_messages) > len(cli_messages)" in handler
@@ -1,11 +1,12 @@
 """Tests for backend tool-call summary extraction used by WebUI session persistence."""
+import json
 import pathlib
 import sys

 REPO_ROOT = pathlib.Path(__file__).parent.parent.resolve()
 sys.path.insert(0, str(REPO_ROOT))

-from api.streaming import _extract_tool_calls_from_messages
+from api.streaming import _extract_tool_calls_from_messages, _tool_result_snippet


 def test_extract_tool_calls_from_openai_message_linkage():
@@ -32,6 +33,64 @@ def test_extract_tool_calls_from_openai_message_linkage():
    assert result[0]["snippet"] == "file.txt"


+def test_tool_result_snippet_allows_frontend_show_more_threshold_but_stays_bounded():
+    """Persisted snippets should be long enough for frontend Show more but capped."""
+    medium_output = "m" * 1200
+    huge_output = "h" * 5000
+
+    medium_snippet = _tool_result_snippet(json.dumps({"output": medium_output}))
+    huge_snippet = _tool_result_snippet(json.dumps({"output": huge_output}))
+
+    assert len(medium_snippet) == 1200
+    assert len(medium_snippet) > 800
+    assert len(huge_snippet) == 4000
+
+
+def test_extract_tool_calls_persists_show_more_sized_snippets_with_bounded_cap():
+    """Tool-call summaries should store >800-char snippets without growing unbounded."""
+    long_output = "x" * 1200
+    huge_output = "y" * 5000
+    messages = [
+        {
+            "role": "assistant",
+            "content": "",
+            "tool_calls": [
+                {
+                    "id": "call-long",
+                    "function": {
+                        "name": "read_file",
+                        "arguments": '{"path":"/tmp/medium.log"}',
+                    },
+                },
+                {
+                    "id": "call-huge",
+                    "function": {
+                        "name": "terminal",
+                        "arguments": '{"command":"yes"}',
+                    },
+                },
+            ],
+        },
+        {
+            "role": "tool",
+            "tool_call_id": "call-long",
+            "content": json.dumps({"output": long_output}),
+        },
+        {
+            "role": "tool",
+            "tool_call_id": "call-huge",
+            "content": json.dumps({"output": huge_output}),
+        },
+    ]
+
+    result = _extract_tool_calls_from_messages(messages)
+
+    assert len(result) == 2
+    assert len(result[0]["snippet"]) == 1200
+    assert len(result[0]["snippet"]) > 800
+    assert len(result[1]["snippet"]) == 4000
+
+
 def test_extract_tool_calls_falls_back_to_live_progress_when_ids_missing():
    messages = [
        {"role": "user", "content": "write spec"},
@@ -8,6 +8,7 @@ from pathlib import Path
 REPO = Path(__file__).resolve().parent.parent
 STREAMING_PY = (REPO / "api" / "streaming.py").read_text(encoding="utf-8")
 MESSAGES_JS = (REPO / "static" / "messages.js").read_text(encoding="utf-8")
+ROUTES_PY = (REPO / "api" / "routes.py").read_text(encoding="utf-8")
 UI_JS = (REPO / "static" / "ui.js").read_text(encoding="utf-8")
 CSS = (REPO / "static" / "style.css").read_text(encoding="utf-8")

@@ -61,3 +62,29 @@ def test_ui_formats_and_renders_turn_duration_in_footer_and_activity_summary():
    assert ".msg-duration-inline" in CSS and ".tool-call-group-duration" in CSS, (
        "Duration UI should have explicit CSS hooks for the footer chip and compact activity summary."
    )
+
+
+def test_active_compact_activity_elapsed_timer_uses_persisted_start_time():
+    assert '"pending_started_at": s.pending_started_at' in ROUTES_PY, (
+        "/api/chat/start should return the persisted pending_started_at timestamp "
+        "so the live timer starts from backend/session truth."
+    )
+    assert "startData.pending_started_at" in MESSAGES_JS, (
+        "send() should copy chat-start pending_started_at into S.session before "
+        "attaching the live stream."
+    )
+    assert "function _formatActiveElapsedTimer" in UI_JS and "padStart(2,'0')" in UI_JS, (
+        "ui.js should format the running timer in MM:SS form."
+    )
+    assert "data-turn-started-at" in UI_JS and "data-active-turn-elapsed" in UI_JS, (
+        "Live compact Activity groups need stable start-time and active-elapsed "
+        "hooks for browser QA and reconnect/rerender safety."
+    )
+    assert "Working " in UI_JS, (
+        "The in-progress Activity summary should distinguish the live counter "
+        "from the settled 'Done in …' duration."
+    )
+    assert "setInterval" in UI_JS and "_clearActivityElapsedTimer" in UI_JS, (
+        "The active elapsed label should tick while running and clear its interval "
+        "on terminal/error/session-switch cleanup paths."
+    )