Merge remote-tracking branch 'origin/master' into feat/activity-feed-default-open

# Conflicts:
#	CHANGELOG.md

Author: AJV20

CHANGELOG.md

@@ -8,6 +8,22 @@
 - Activity disclosures now have a Settings → Appearance option to expand the activity feed by default, while still honoring per-turn manual collapse/expand choices.
 - Live Activity waiting rows now explain the agent's current step instead of stopping at `Waiting on model`, including a prompt/context review hint before tools run and the latest tool-derived action while the model chooses the next step.
 
+## [v0.51.157] — 2026-05-28 — Release EC (stage-batch39 — 5-PR mixed-risk cleanup: gateway prefill forward + prefill budget + compressed-continuation sidebar + browser-transcript memory guidance + reasoning max parity)
+
+### Added
+
+- The reasoning-effort selector now offers a `max` level, matching the agent's `hermes_constants.VALID_REASONING_EFFORTS`. This restores parity with the underlying set (the WebUI mirror previously stopped at `xhigh`) so providers such as Anthropic that support the `max` thinking level are selectable from the composer dropdown and the `/reasoning` command.
+
+### Changed
+
+- WebUI's browser-session surface prompt now explicitly tells agents not to dump browser transcripts into external notes or durable memory by default; it limits saving to explicit captures and clearly reusable durable signals such as preferences, decisions, blockers, and runbook-worthy workflows.
+
+### Fixed
+
+- Gateway-backed WebUI chat now forwards configured prefill/session-recall context and a compact WebUI session-context block into delegated Gateway turns, so browser sessions retain note recall, connected-platform awareness, and delivery hints instead of sending only the latest user message. If the dynamic prefill script fails, WebUI falls back to the configured static router prefill when available.
+- Oversized WebUI startup prefill payloads now respect a configurable context budget (`webui_prefill_context_max_chars` / `HERMES_WEBUI_PREFILL_CONTEXT_MAX_CHARS`, default 12,000 chars). When a dynamic prefill script exceeds the budget and a compact static prefill file is configured, WebUI falls back to the compact file; otherwise it injects a small retrieval instruction instead of dumping the full note/body payload into every new chat.
+- Sidebar now keeps the newest active continuation visible when it has more recent activity than an older fuller pre-compression snapshot in the same lineage. Adds lineage-aware dedupe for WebUI-origin state-db projections, restores normal context-only turns into the visible transcript after compression while preserving order, and recognizes `[Session Arc Summary]` as a compression marker so it isn't backfilled into the chat transcript.
+
 ## [v0.51.156] — 2026-05-28 — Release EB (stage-batch38 — 2-PR Tier B cleanup: WebUI request/runtime hardening + chat-start provider fallback)
 
 ### Fixed

api/compression_anchor.py

@@ -68,6 +68,7 @@ def is_context_compression_marker(message):
         text.startswith("[context compaction")
         or text.startswith("context compaction")
         or text.startswith("[your active task list was preserved across context compression]")
+        or text.startswith("[session arc summary")
     )
 
 

api/config.py

@@ -2073,7 +2073,7 @@ def get_effective_default_model(config_data: dict | None = None) -> str:
 # Mirrors hermes_constants.parse_reasoning_effort so WebUI can validate without
 # importing from the agent tree (which may not be installed).  Any drift here
 # will show up in the shared test suite since both sides accept the same set.
-VALID_REASONING_EFFORTS = ("minimal", "low", "medium", "high", "xhigh")
+VALID_REASONING_EFFORTS = ("minimal", "low", "medium", "high", "xhigh", "max")
 
 
 def parse_reasoning_effort(effort):

api/gateway_chat.py

@@ -223,6 +223,22 @@ def put_gateway_event(event, data):
         from api.config import get_config  # imported lazily to avoid config-cycle churn
 
         cfg = get_config()
+        try:
+            from api.streaming import (
+                _load_webui_prefill_context,
+                _prefill_messages_with_webui_context,
+                _public_prefill_context_status,
+            )
+
+            prefill_context = _load_webui_prefill_context(cfg)
+            prefill_messages = _prefill_messages_with_webui_context(prefill_context, cfg)
+            put_gateway_event("context_status", {
+                "session_id": session_id,
+                "prefill": _public_prefill_context_status(prefill_context),
+            })
+        except Exception:
+            logger.debug("Failed to load WebUI gateway prefill context", exc_info=True)
+            prefill_messages = []
         base_url = _gateway_base_url(cfg)
         api_key = _gateway_api_key()
         url = f"{base_url}/v1/chat/completions"
@@ -248,7 +264,7 @@ def put_gateway_event(event, data):
         body = {
             "model": model or "default",
             "stream": True,
-            "messages": [{"role": "user", "content": message_content}],
+            "messages": [*prefill_messages, {"role": "user", "content": message_content}],
         }
         if model_provider:
             body["provider"] = model_provider

api/models.py

@@ -2495,6 +2495,14 @@ def _prefer_fuller_snapshots_for_sidebar(sessions: list[dict]) -> list[dict]:
         if _sidebar_message_count(best_snapshot) <= best_visible_count:
             continue
 
+        newest_visible_ts = max(_session_sort_timestamp(session) for session in visible)
+        snapshot_ts = _session_sort_timestamp(best_snapshot)
+        # Keep the active continuation visible when it has newer activity than
+        # the archived snapshot. A fuller snapshot can still be older than a
+        # continuation that contains the latest turns after compression.
+        if newest_visible_ts > snapshot_ts:
+            continue
+
         snapshot_ids_to_show.add(str(best_snapshot.get('session_id')))
         continuation_ids_to_hide.update(
             str(session.get('session_id'))

api/routes.py

@@ -2457,6 +2457,43 @@ def _normalize_sidebar_source_flags(session: dict) -> dict:
     return normalized
 
 
+def _session_source_is_webui(session: dict) -> bool:
+    """Return True for state.db/sidebar rows that describe WebUI-origin sessions."""
+    if not isinstance(session, dict):
+        return False
+    for key in ("source_tag", "raw_source", "session_source", "source"):
+        if str(session.get(key) or "").strip().lower() == "webui":
+            return True
+    return False
+
+
+def _session_lineage_ids(session: dict) -> set[str]:
+    """Return known ids that identify one logical sidebar lineage."""
+    if not isinstance(session, dict):
+        return set()
+    ids: set[str] = set()
+    for key in ("session_id", "_lineage_root_id", "_lineage_tip_id"):
+        value = session.get(key)
+        if value:
+            ids.add(str(value))
+    return ids
+
+
+def _is_duplicate_webui_state_projection(session: dict, represented_webui_ids: set[str]) -> bool:
+    """Return True when a state.db row is only a duplicate WebUI-origin projection.
+
+    The "Show non-WebUI sessions" toggle should add external/agent-owned
+    conversations, not make WebUI compression continuations appear only when the
+    external-session bridge is enabled. WebUI-origin state.db rows are still
+    useful metadata sidecars, but if any id in their compression lineage is
+    already represented by WebUI session JSON, they should not be injected as an
+    additive external row.
+    """
+    if not _session_source_is_webui(session):
+        return False
+    return bool(_session_lineage_ids(session) & represented_webui_ids)
+
+
 CLI_VISIBLE_SESSION_CAP = 20
 
 
@@ -4518,9 +4555,17 @@ def handle_get(handler, parsed) -> bool:
                 # Apply the same CLI visibility semantics to imported local copies so
                 # low-value imported artifacts do not leak into the sidebar.
                 webui_sessions = [s for s in webui_sessions if is_cli_session_row_visible(s)]
-                webui_ids = {s["session_id"] for s in webui_sessions}
+                represented_webui_ids = set()
+                for s in webui_sessions:
+                    represented_webui_ids.update(_session_lineage_ids(s))
                 from api.models import _hide_from_default_sidebar as _cron_hide
-                deduped_cli = [s for s in cli if s["session_id"] not in webui_ids and is_cli_session_row_visible(s) and not _cron_hide(s)]
+                deduped_cli = [
+                    s for s in cli
+                    if s["session_id"] not in represented_webui_ids
+                    and not _is_duplicate_webui_state_projection(s, represented_webui_ids)
+                    and is_cli_session_row_visible(s)
+                    and not _cron_hide(s)
+                ]
             else:
                 diag.stage("filter_webui_sessions")
                 webui_sessions = [s for s in webui_sessions if not _is_cli_session_for_settings(s)]