test: add fire_and_forget to API chat webhook stub

docker-compose.yml injects FASTEMBED_CACHE_PATH=${FASTEMBED_CACHE_PATH:-},
which sets the variable to an empty string when the host has not defined it.
FASTEMBED_CACHE_DIR used os.getenv("FASTEMBED_CACHE_PATH", default), and
os.getenv only returns the default when the variable is ABSENT -- so the empty
value won and FASTEMBED_CACHE_DIR became "". os.makedirs("") then raised
[Errno 2] No such file or directory: '', FastEmbed failed to initialise, and
every vector feature (RAG, semantic memory, tool index) silently degraded on
the default Docker stack.

Treat an empty value like an absent one via `os.getenv(...) or default`.
Add a regression test covering the empty, unset, and explicit cases.

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

companion/routes.py

@@ -5,8 +5,9 @@ offers and pair to it, without duplicating any LLM logic.
 
 Auth is enforced globally by AuthMiddleware (app.py), so reaching a handler here
 means the caller is authenticated by either a cookie session or a Bearer `ody_`
-API token. The read endpoints (ping/info/models) accept either; the pairing
-endpoints are admin-cookie only.
+API token. Ping/info accept either credential type, models requires a chat-
+scoped API token for bearer callers, and the pairing endpoints are admin-cookie
+only.
 
 Pairing CSRF posture: minting happens ONLY on POST. The session cookie is
 SameSite=Lax (routes/auth_routes.py), which a browser does not send on a
@@ -18,7 +19,7 @@ on a GET would be unsafe (Lax cookies ride top-level GET navigations), so GET
 
 import html
 
-from fastapi import APIRouter, Request
+from fastapi import APIRouter, HTTPException, Request
 from fastapi.responses import HTMLResponse
 
 from core.middleware import require_admin
@@ -52,6 +53,18 @@ def owner_can_see(row_owner, owner) -> bool:
     return row_owner is None or row_owner == owner
 
 
+def require_models_scope(request: Request) -> None:
+    """Require the companion chat scope for bearer-token model inventory."""
+    if not getattr(request.state, "api_token", False):
+        return
+    scopes = getattr(request.state, "api_token_scopes", None) or []
+    if isinstance(scopes, str):
+        scopes = [scope.strip() for scope in scopes.split(",")]
+    scope_set = {str(scope).strip() for scope in scopes if str(scope).strip()}
+    if _pairing.COMPANION_SCOPE not in scope_set:
+        raise HTTPException(403, "API token requires chat scope")
+
+
 def mint_pairing_token(owner: str, invalidate=None) -> tuple[str, str]:
     """Mint a pairing token AND invalidate the auth middleware's in-memory token
     cache, so the new token is accepted on the very next request without a server
@@ -103,6 +116,7 @@ def setup_companion_routes() -> APIRouter:
         rows -- the same rule as owner_filter. Read-only; never returns api_key
         material.
         """
+        require_models_scope(request)
         import json as _json
 
         from core.database import SessionLocal, ModelEndpoint

mcp_servers/memory_server.py

@@ -6,6 +6,7 @@ Imports MemoryManager and MemoryVectorStore from the Odysseus codebase.
 """
 
 import asyncio
+import os
 import sys
 import time
 from pathlib import Path
@@ -23,6 +24,55 @@ _memory_manager = None
 _memory_vector = None
 _initialized = False
 
+_OWNER_ENV_KEYS = ("ODYSSEUS_MCP_MEMORY_OWNER", "ODYSSEUS_MEMORY_OWNER")
+_OWNER_SCOPE_ERROR = (
+    "Error: Memory MCP owner is not configured for an owner-scoped memory store. "
+    "Set ODYSSEUS_MCP_MEMORY_OWNER for this server or use the owner-aware native memory tool."
+)
+
+
+def _configured_owner() -> str | None:
+    for key in _OWNER_ENV_KEYS:
+        owner = os.environ.get(key, "").strip()
+        if owner:
+            return owner
+    return None
+
+
+def _entry_owner(entry: dict) -> str | None:
+    owner = entry.get("owner")
+    if owner is None:
+        return None
+    owner_text = str(owner).strip()
+    return owner_text or None
+
+
+def _owner_scoped_store(entries: list[dict]) -> bool:
+    return any(_entry_owner(entry) for entry in entries if isinstance(entry, dict))
+
+
+def _scope_entries() -> tuple[str | None, list[dict], list[dict], str | None]:
+    """Return configured owner, all entries, visible entries, and optional error."""
+    entries = _memory_manager.load_all()
+    owner = _configured_owner()
+    if owner is None and _owner_scoped_store(entries):
+        return None, entries, [], _OWNER_SCOPE_ERROR
+    if owner is None:
+        visible = [
+            entry for entry in entries
+            if isinstance(entry, dict) and _entry_owner(entry) is None
+        ]
+    else:
+        visible = [
+            entry for entry in entries
+            if isinstance(entry, dict) and _entry_owner(entry) == owner
+        ]
+    return owner, entries, visible, None
+
+
+def _text_result(text: str) -> list[TextContent]:
+    return [TextContent(type="text", text=text)]
+
 
 def _ensure_init():
     """Lazy-init memory managers on first use."""
@@ -75,24 +125,26 @@ async def list_tools() -> list[Tool]:
 @server.call_tool()
 async def call_tool(name: str, arguments: dict) -> list[TextContent]:
     if name != "manage_memory":
-        return [TextContent(type="text", text=f"Unknown tool: {name}")]
+        return _text_result(f"Unknown tool: {name}")
 
     _ensure_init()
     if not _memory_manager:
-        return [TextContent(type="text", text="Error: Memory manager not available")]
+        return _text_result("Error: Memory manager not available")
 
     action = arguments.get("action", "")
 
     if action == "list":
         category_filter = arguments.get("category", "")
-        memories = _memory_manager.load()
+        _owner, _all_memories, memories, scope_error = _scope_entries()
+        if scope_error:
+            return _text_result(scope_error)
         if category_filter:
             memories = [m for m in memories if m.get("category", "").lower() == category_filter.lower()]
         if not memories:
             msg = "No memories found"
             if category_filter:
                 msg += f" in category '{category_filter}'"
-            return [TextContent(type="text", text=msg + ".")]
+            return _text_result(msg + ".")
 
         lines = [f"Found {len(memories)} memory entries:\n"]
         for m in memories:
@@ -102,15 +154,17 @@ async def call_tool(name: str, arguments: dict) -> list[TextContent]:
             if len(text) > 150:
                 text = text[:150] + "..."
             lines.append(f"- [{cat}] `{mid}` — {text}")
-        return [TextContent(type="text", text="\n".join(lines))]
+        return _text_result("\n".join(lines))
 
     elif action == "add":
         text = arguments.get("text", "")
         category = arguments.get("category", "fact")
         if not text:
-            return [TextContent(type="text", text="Error: Memory text cannot be empty")]
-        entry = _memory_manager.add_entry(text, source="ai_agent", category=category)
-        memories = _memory_manager.load_all()
+            return _text_result("Error: Memory text cannot be empty")
+        owner, memories, _visible, scope_error = _scope_entries()
+        if scope_error:
+            return _text_result(scope_error)
+        entry = _memory_manager.add_entry(text, source="ai_agent", category=category, owner=owner)
         memories.append(entry)
         _memory_manager.save(memories)
         if _memory_vector and _memory_vector.healthy:
@@ -118,25 +172,28 @@ async def call_tool(name: str, arguments: dict) -> list[TextContent]:
                 _memory_vector.add(entry["id"], text)
             except Exception:
                 pass
-        return [TextContent(type="text", text=f"Memory added: [{category}] {text} (id: {entry['id'][:8]})")]
+        return _text_result(f"Memory added: [{category}] {text} (id: {entry['id'][:8]})")
 
     elif action == "edit":
         memory_id = arguments.get("memory_id", "")
         new_text = arguments.get("text", "")
         if not memory_id or not new_text:
-            return [TextContent(type="text", text="Error: edit needs memory_id and text")]
-        memories = _memory_manager.load_all()
-        found = False
+            return _text_result("Error: edit needs memory_id and text")
+        _owner, memories, visible, scope_error = _scope_entries()
+        if scope_error:
+            return _text_result(scope_error)
         full_id = None
-        for m in memories:
+        for m in visible:
             if m.get("id", "").startswith(memory_id):
-                m["text"] = new_text
-                m["timestamp"] = int(time.time())
-                found = True
                 full_id = m["id"]
                 break
-        if not found:
-            return [TextContent(type="text", text=f"Error: Memory '{memory_id}' not found")]
+        if not full_id:
+            return _text_result(f"Error: Memory '{memory_id}' not found")
+        for m in memories:
+            if m.get("id") == full_id:
+                m["text"] = new_text
+                m["timestamp"] = int(time.time())
+                break
         _memory_manager.save(memories)
         if _memory_vector and _memory_vector.healthy and full_id:
             try:
@@ -144,24 +201,26 @@ async def call_tool(name: str, arguments: dict) -> list[TextContent]:
                 _memory_vector.add(full_id, new_text)
             except Exception:
                 pass
-        return [TextContent(type="text", text=f"Memory updated: {new_text}")]
+        return _text_result(f"Memory updated: {new_text}")
 
     elif action == "delete":
         memory_id = arguments.get("memory_id", "")
         if not memory_id:
-            return [TextContent(type="text", text="Error: delete needs memory_id")]
-        memories = _memory_manager.load_all()
+            return _text_result("Error: delete needs memory_id")
+        _owner, memories, visible, scope_error = _scope_entries()
+        if scope_error:
+            return _text_result(scope_error)
         full_id = None
         deleted_text = ""
         deleted_category = ""
-        for m in memories:
+        for m in visible:
             if m.get("id", "").startswith(memory_id):
                 full_id = m["id"]
                 deleted_text = m.get("text", "")
                 deleted_category = m.get("category", "")
                 break
         if not full_id:
-            return [TextContent(type="text", text=f"Error: Memory '{memory_id}' not found")]
+            return _text_result(f"Error: Memory '{memory_id}' not found")
         memories = [m for m in memories if m.get("id") != full_id]
         _memory_manager.save(memories)
         if _memory_vector and _memory_vector.healthy and full_id:
@@ -171,30 +230,32 @@ async def call_tool(name: str, arguments: dict) -> list[TextContent]:
                 pass
         cat = f"[{deleted_category}] " if deleted_category else ""
         snippet = deleted_text if len(deleted_text) <= 120 else deleted_text[:117] + "..."
-        return [TextContent(type="text", text=f"Memory deleted: {cat}{snippet} (id: {memory_id})")]
+        return _text_result(f"Memory deleted: {cat}{snippet} (id: {memory_id})")
 
     elif action == "search":
         query = arguments.get("text", "")
         if not query:
-            return [TextContent(type="text", text="Error: search needs text (query)")]
-        memories = _memory_manager.load()
+            return _text_result("Error: search needs text (query)")
+        _owner, _all_memories, memories, scope_error = _scope_entries()
+        if scope_error:
+            return _text_result(scope_error)
         if hasattr(_memory_manager, 'get_relevant_memories'):
             results = _memory_manager.get_relevant_memories(query, memories, threshold=0.05, max_items=20)
         else:
             query_lower = query.lower()
             results = [m for m in memories if query_lower in m.get("text", "").lower()][:20]
         if not results:
-            return [TextContent(type="text", text=f"No memories found matching '{query}'.")]
+            return _text_result(f"No memories found matching '{query}'.")
         lines = [f"Found {len(results)} matching memories:\n"]
         for m in results:
             cat = m.get("category", "fact")
             mid = m.get("id", "?")[:8]
             text = m.get("text", "")
             lines.append(f"- [{cat}] `{mid}` — {text}")
-        return [TextContent(type="text", text="\n".join(lines))]
+        return _text_result("\n".join(lines))
 
     else:
-        return [TextContent(type="text", text=f"Error: Unknown action '{action}'. Use: list, add, edit, delete, search")]
+        return _text_result(f"Error: Unknown action '{action}'. Use: list, add, edit, delete, search")
 
 
 async def run():

routes/chat_helpers.py

@@ -14,7 +14,7 @@ from core.database import Session as DBSession, ModelEndpoint
 from src.llm_core import normalize_model_id
 from src.endpoint_resolver import normalize_base
 from src.context_compactor import maybe_compact, trim_for_context
-from src.auth_helpers import get_current_user
+from src.auth_helpers import effective_user
 from src.prompt_security import untrusted_context_message
 from routes.prefs_routes import _load_for_user as load_prefs_for_user
 
@@ -78,7 +78,7 @@ def _enforce_chat_privileges(request, sess) -> None:
     which means unrestricted allowed_models / zero cap -> no-op for them.
     """
     try:
-        user = get_current_user(request)
+        user = effective_user(request)
     except Exception:
         user = None
     if not user:
@@ -346,11 +346,11 @@ def add_user_message(sess, chat_handler, preprocessed: PreprocessedMessage, inco
 def fire_message_event(request, webhook_manager, session_id: str, sess, message: str, compare_mode: bool = False):
     """Fire webhook and event_bus events for a new user message."""
     if webhook_manager and not compare_mode:
-        asyncio.create_task(webhook_manager.fire("chat.message", {
+        webhook_manager.fire_and_forget("chat.message", {
             "session_id": session_id, "model": sess.model, "message": message[:2000],
-        }))
+        })
     from src.event_bus import fire_event
-    user = get_current_user(request)
+    user = effective_user(request)
     fire_event("message_sent", user)
 
 
@@ -577,7 +577,7 @@ async def build_chat_context(
         fire_message_event(request, webhook_manager, session_id, sess, message, compare_mode)
 
     # Resolve user prefs
-    user = get_current_user(request)
+    user = effective_user(request)
     uprefs = load_prefs_for_user(user)
 
     # Memory enabled?
@@ -1120,10 +1120,10 @@ def run_post_response_tasks(
 
     # Webhook
     if webhook_manager and not compare_mode:
-        asyncio.create_task(webhook_manager.fire("chat.completed", {
+        webhook_manager.fire_and_forget("chat.completed", {
             "session_id": session_id, "model": sess.model,
             "user_message": message, "response": full_response[:2000],
-        }))
+        })
 
     # Auto-name
     if needs_auto_name(sess.name):

routes/chat_routes.py

@@ -23,7 +23,7 @@ from src.endpoint_resolver import normalize_base as _normalize_base, build_chat_
 from src.session_search import search_session_messages
 from src.prompt_security import untrusted_context_message
 from core.exceptions import SessionNotFoundError
-from src.auth_helpers import get_current_user
+from src.auth_helpers import effective_user, get_current_user
 from routes.session_routes import _verify_session_owner
 from routes.document_helpers import _owner_session_filter
 from core.database import SessionLocal, get_session_mode, set_session_mode
@@ -363,7 +363,7 @@ def setup_chat_routes(
             sess = session_manager.get_session(session)
         except KeyError:
             raise HTTPException(404, f"Session '{session}' not found")
-        owner = get_current_user(request)
+        owner = effective_user(request)
         if _clear_orphaned_session_endpoint(sess, owner=owner):
             raise HTTPException(400, "Selected model endpoint was removed. Pick another model in Settings.")
 
@@ -603,7 +603,7 @@ def setup_chat_routes(
             # but BEFORE loading. Prevents cross-user session hijack.
             _verify_session_owner(request, session)
             sess = session_manager.get_session(session)
-            owner = get_current_user(request)
+            owner = effective_user(request)
             if _clear_orphaned_session_endpoint(sess, owner=owner):
                 raise HTTPException(400, "Selected model endpoint was removed. Pick another model in Settings.")
             # Issue #587: picker shows a model from the endpoint cache but
@@ -634,7 +634,7 @@ def setup_chat_routes(
         _enforce_chat_privileges(request, sess)
 
         # Ensure session has auth headers
-        resolve_session_auth(sess, session, owner=get_current_user(request))
+        resolve_session_auth(sess, session, owner=effective_user(request))
 
         # Check for research_pending BEFORE mode persist overwrites it
         do_research = str(use_research).lower() == "true"
@@ -1485,7 +1485,7 @@ def setup_chat_routes(
         if not q or not q.strip():
             return []
 
-        _user = get_current_user(request)
+        _user = effective_user(request)
         return [
             result.to_dict()
             for result in search_session_messages(

routes/cookbook_helpers.py

@@ -505,6 +505,8 @@ def _cached_model_scan_script(model_dirs: list[str] | None = None, add_hf_cache:
         "    if u.startswith('KB'): return int(n * 1024)",
         "    return int(n)",
         "def scan_ollama():",
+        "    if any(m.get('is_ollama') for m in models): return",
+        "    if os.name == 'nt' and not os.environ.get('ODYSSEUS_ALLOW_OLLAMA_CLI_SCAN'): return",
         "    if not shutil.which('ollama'): return",
         "    try:",
         "        p = subprocess.run(['ollama', 'list'], stdout=subprocess.PIPE, stderr=subprocess.DEVNULL, text=True, timeout=6)",
@@ -535,8 +537,8 @@ def _cached_model_scan_script(model_dirs: list[str] | None = None, add_hf_cache:
         "            models.append({'repo_id':name,'size_bytes':size_bytes,'nb_files':1,'has_incomplete':False,'path':'ollama','backend':'ollama','is_ollama':True})",
         "        return",
         "for _hf_cache in hf_cache_paths(): scan_hf(_hf_cache)",
-        "scan_ollama()",
         "scan_ollama_api()",
+        "scan_ollama()",
     ]
     for model_dir in model_dirs or []:
         lines.append(f"scan_dir(os.path.expanduser({model_dir!r}))")

routes/session_routes.py

@@ -11,7 +11,7 @@ from core.session_manager import SessionManager
 from core.models import ChatMessage
 from src.request_models import SessionResponse
 from core.database import Session as DbSession, SessionLocal, Document, GalleryImage, utcnow_naive
-from src.auth_helpers import get_current_user, effective_user, _auth_disabled, owner_filter
+from src.auth_helpers import effective_user, _auth_disabled, owner_filter
 from src.session_actions import is_session_recently_active
 
 
@@ -328,7 +328,7 @@ def setup_session_routes(session_manager: SessionManager, config: dict, webhook_
         endpoint_id: str = Form(""),
     ):
         skip_val = str(skip_validation).lower() == "true"
-        user = get_current_user(request)
+        user = effective_user(request)
         endpoint_api_key = ""
         endpoint_base_url = ""
         _reject_raw_endpoint_url_for_non_admin(request, user, endpoint_id, endpoint_url)
@@ -477,7 +477,7 @@ def setup_session_routes(session_manager: SessionManager, config: dict, webhook_
                 db.close()
         # Switch model/endpoint mid-session
         if model is not None and endpoint_url is not None:
-            user = get_current_user(request)
+            user = effective_user(request)
             _reject_raw_endpoint_url_for_non_admin(request, user, endpoint_id, endpoint_url)
             endpoint_api_key = ""
             endpoint_base_url = ""

routes/upload_routes.py

@@ -7,7 +7,7 @@ from fastapi import APIRouter, Request, File, UploadFile, HTTPException
 from typing import List
 import logging
 from core.middleware import require_admin
-from src.auth_helpers import get_current_user
+from src.auth_helpers import effective_user
 from src.upload_handler import count_recent_uploads
 
 logger = logging.getLogger(__name__)
@@ -78,7 +78,7 @@ def setup_upload_routes(upload_handler):
         
         for u in files:
             try:
-                meta = upload_handler.save_upload(u, client_ip, owner=get_current_user(request))
+                meta = upload_handler.save_upload(u, client_ip, owner=effective_user(request))
                 out.append({
                     "id": meta["id"],
                     "name": meta["name"],
@@ -138,7 +138,7 @@ def setup_upload_routes(upload_handler):
                 original_name = info.get("name", file_id)
         auth_mgr = getattr(request.app.state, "auth_manager", None)
         auth_configured = bool(auth_mgr and auth_mgr.is_configured)
-        current_user = get_current_user(request)
+        current_user = effective_user(request)
         file_owner = info.get("owner") if info else None
         if auth_configured:
             if not current_user:
@@ -204,7 +204,7 @@ def setup_upload_routes(upload_handler):
         info = _load_upload_info(file_id)
         auth_mgr = getattr(request.app.state, "auth_manager", None)
         auth_configured = bool(auth_mgr and auth_mgr.is_configured)
-        current_user = get_current_user(request)
+        current_user = effective_user(request)
         file_owner = info.get("owner") if info else None
         if auth_configured:
             if not current_user:
@@ -247,7 +247,7 @@ def setup_upload_routes(upload_handler):
             raise HTTPException(404, "File not found")
         auth_mgr = getattr(request.app.state, "auth_manager", None)
         auth_configured = bool(auth_mgr and auth_mgr.is_configured)
-        current_user = get_current_user(request)
+        current_user = effective_user(request)
         file_owner = info.get("owner")
         if auth_configured:
             if not current_user:

routes/webhook_routes.py

@@ -1,6 +1,5 @@
 """Webhook, API Token, and sync chat routes."""
 
-import asyncio
 import uuid
 import logging
 from typing import Optional
@@ -385,10 +384,10 @@ def setup_webhook_routes(
         sess.add_message(ChatMessage("assistant", reply))
         session_manager.save_sessions()
 
-        asyncio.create_task(webhook_manager.fire("chat.completed", {
+        webhook_manager.fire_and_forget("chat.completed", {
             "session_id": session_id, "model": sess.model,
             "user_message": message[:2000], "response": reply[:2000],
-        }))
+        })
 
         return {"response": reply, "session_id": session_id, "model": sess.model}
 

services/search/content.py

@@ -15,7 +15,7 @@ from urllib.parse import urljoin, urlparse
 import httpx
 from bs4 import BeautifulSoup
 
-from src.constants import WEB_FETCH_SOFT_MAX_BYTES, WEB_FETCH_HARD_MAX_BYTES
+from src.constants import WEB_FETCH_SOFT_MAX_BYTES, WEB_FETCH_HARD_MAX_BYTES, WEB_FETCH_USER_AGENT
 
 from .analytics import RateLimitError, error_logger
 from .cache import (
@@ -369,7 +369,7 @@ def fetch_webpage_content(url: str, timeout: int = 5, retry_attempt: int = 0,
     # Fetch
     try:
         headers = {
-            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36",
+            "User-Agent": WEB_FETCH_USER_AGENT,
             "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
             "Accept-Language": "en-US,en;q=0.5",
             # identity so the streamed size cap in _get_public_url stays honest

services/search/providers.py

@@ -9,7 +9,7 @@ from urllib.parse import urljoin, urlparse, parse_qs
 import httpx
 from bs4 import BeautifulSoup
 
-from src.constants import SEARXNG_INSTANCE, REQUEST_TIMEOUT
+from src.constants import SEARXNG_INSTANCE, REQUEST_TIMEOUT, WEB_FETCH_USER_AGENT
 from .analytics import RateLimitError, error_logger
 from .query import build_enhanced_query
 
@@ -138,7 +138,7 @@ def searxng_search_api(query: str, count: Optional[int] = None, categories: str
     count = count if count is not None else _get_result_count()
     instance = _get_search_instance()
     api_key = ""
-    headers = {"User-Agent": "Mozilla/5.0"}
+    headers = {"User-Agent": WEB_FETCH_USER_AGENT}
     if api_key:
         headers["Authorization"] = f"Bearer {api_key}"
     # News/fresh queries do badly in the 'general' category — it favours
@@ -250,7 +250,7 @@ def searxng_search(query, max_results=10):
     """Search using SearXNG instance - parsing HTML."""
     instance = _get_search_instance()
     api_key = ""
-    req_headers = {"User-Agent": "Mozilla/5.0"}
+    req_headers = {"User-Agent": WEB_FETCH_USER_AGENT}
     if api_key:
         req_headers["Authorization"] = f"Bearer {api_key}"
     try:
@@ -389,7 +389,7 @@ def duckduckgo_search(query: str, count: Optional[int] = None, time_filter: Opti
             response = httpx.get(
                 "https://html.duckduckgo.com/html/",
                 params={"q": query, "kp": _safesearch_for("duckduckgo_html")},
-                headers={"User-Agent": "Mozilla/5.0"},
+                headers={"User-Agent": WEB_FETCH_USER_AGENT},
                 timeout=REQUEST_TIMEOUT,
             )
             response.raise_for_status()

src/constants.py

@@ -57,7 +57,13 @@ MEMORY_VECTORS_DIR = os.path.join(DATA_DIR, "memory_vectors")
 
 # Paths with an intentional dedicated env override, defaulting under DATA_DIR.
 MAIL_ATTACHMENTS_DIR = os.getenv("ODYSSEUS_MAIL_ATTACHMENTS_DIR", os.path.join(DATA_DIR, "mail-attachments"))
-FASTEMBED_CACHE_DIR = os.getenv("FASTEMBED_CACHE_PATH", os.path.join(DATA_DIR, "fastembed_cache"))
+# `or` (not os.getenv's default arg) so a PRESENT-but-EMPTY value falls back to
+# the default. docker-compose.yml injects `FASTEMBED_CACHE_PATH=${FASTEMBED_CACHE_PATH:-}`,
+# which sets the var to "" when the host hasn't defined it. os.getenv(name, default)
+# only returns the default when the var is ABSENT, so the empty string would win →
+# os.makedirs("") raises [Errno 2] No such file or directory: '' → FastEmbed fails to
+# init and all vector features (RAG, semantic memory, tool index) silently degrade.
+FASTEMBED_CACHE_DIR = os.getenv("FASTEMBED_CACHE_PATH") or os.path.join(DATA_DIR, "fastembed_cache")
 
 # Agent tool output limits (single source of truth — imported by tool_execution.py,
 # tool_implementations.py, agent_tools.py, and any other module that needs them)
@@ -78,6 +84,13 @@ MAX_CONTEXT_MESSAGES = 90
 REQUEST_TIMEOUT = 20
 OPENAI_COMPAT_PATH = "/v1/chat/completions"
 
+# Outbound UA for web_fetch / web_search scraping; common desktop UA so pages serve normal HTML.
+WEB_FETCH_USER_AGENT = os.environ.get(
+    "WEB_FETCH_USER_AGENT",
+    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+    "(KHTML, like Gecko) Chrome/148.0.0.0 Safari/537.36",
+)
+
 # Environment variables with defaults
 DEFAULT_HOST = os.getenv("LLM_HOST", "localhost")
 LLM_HOSTS = [h.strip() for h in os.getenv("LLM_HOSTS", "").split(",") if h.strip()]

src/task_scheduler.py

@@ -236,6 +236,29 @@ def _digest_windows(now):
     ]
 
 
+def _checkin_calendar_events(db, owner, start, end):
+    """Calendar events in [start, end] for ONE owner, for the check-in digest.
+
+    Ownership lives on CalendarCal.owner; events inherit it via calendar_id.
+    The digest query had no owner scope, so it pulled EVERY user's events into
+    one user's check-in (a cross-tenant leak of summaries/locations). Scope it
+    by joining CalendarCal, mirroring routes/calendar_routes.list_events.
+    """
+    from core.database import CalendarEvent as _CE, CalendarCal as _CC
+    return (
+        db.query(_CE)
+        .join(_CC, _CE.calendar_id == _CC.id)
+        .filter(
+            _CC.owner == owner,
+            _CE.dtstart >= start,
+            _CE.dtstart <= end,
+            _CE.status != "cancelled",
+        )
+        .order_by(_CE.dtstart)
+        .all()
+    )
+
+
 class TaskScheduler:
     def __init__(self, session_manager):
         self._session_manager = session_manager
@@ -1127,11 +1150,7 @@ class TaskScheduler:
                     # Strip timezone for naive DB comparison
                     _s = start.replace(tzinfo=None) if start.tzinfo else start
                     _e = end.replace(tzinfo=None) if end.tzinfo else end
-                    evs = _db.query(_CE).filter(
-                        _CE.dtstart >= _s,
-                        _CE.dtstart <= _e,
-                        _CE.status != "cancelled",
-                    ).order_by(_CE.dtstart).all()
+                    evs = _checkin_calendar_events(_db, task.owner, _s, _e)
                     if not evs:
                         continue
                     # Group by importance for richer output

src/tool_implementations.py

@@ -3797,7 +3797,7 @@ async def do_resolve_contact(content: str, owner: Optional[str] = None) -> Dict:
     if not name:
         return {"error": "name is required", "exit_code": 1}
 
-    contacts = {}  # email -> {name, source}
+    contacts = {}  # email_or_phone -> {name, source, phone?}
 
     # 1. CardDAV (Radicale) — structured contacts. Call in-process: a
     # server-side httpx GET to /api/contacts/search carries no session
@@ -3812,10 +3812,18 @@ async def do_resolve_contact(content: str, owner: Optional[str] = None) -> Dict:
             match = q in hay_name or any(q in (e or "").lower() for e in c.get("emails", []))
             if not match:
                 continue
+            has_email = False
             for email in (c.get("emails") or []):
                 email = (email or "").strip().lower()
                 if email and "@" in email:
                     contacts[email] = {"name": c.get("name") or email, "source": "contacts"}
+                    has_email = True
+            # Fall back to phone numbers when the contact has no email address
+            if not has_email:
+                for phone in (c.get("phones") or []):
+                    phone = (phone or "").strip()
+                    if phone:
+                        contacts[phone] = {"name": c.get("name") or phone, "source": "contacts", "phone": phone}
     except Exception:
         pass
 
@@ -3835,8 +3843,11 @@ async def do_resolve_contact(content: str, owner: Optional[str] = None) -> Dict:
         return {"output": f"No contacts found matching '{name}'.", "exit_code": 0}
 
     lines = [f"Contacts matching '{name}':"]
-    for email, info in contacts.items():
-        lines.append(f"- {info['name']} <{email}> ({info['source']})")
+    for key, info in contacts.items():
+        if info.get("phone"):
+            lines.append(f"- {info['name']} — phone: {info['phone']} ({info['source']})")
+        else:
+            lines.append(f"- {info['name']} <{key}> ({info['source']})")
     return {"output": "\n".join(lines), "exit_code": 0}
 
 

src/tool_schemas.py

@@ -1009,7 +1009,7 @@ FUNCTION_TOOL_SCHEMAS = [
         "type": "function",
         "function": {
             "name": "resolve_contact",
-            "description": "Look up a contact's email address by name. Searches CardDAV address book and sent email history. Use when the user says 'message [name]' or 'email [name]' without an email address.",
+            "description": "Look up a contact by name. Searches CardDAV address book and sent email history. Returns email addresses (when available) or phone numbers. Use when the user says 'message [name]', 'email [name]', or asks for someone's contact details.",
             "parameters": {
                 "type": "object",
                 "properties": {

tests/test_api_chat_security.py

@@ -219,6 +219,9 @@ class _WebhookManager:
     async def fire(self, event, payload):
         return None
 
+    def fire_and_forget(self, event, payload):
+        return None
+
 
 def _install_sync_chat_stubs(monkeypatch):
     # FastAPI checks for python_multipart at import time when Form is used;

tests/test_chat_helpers.py

@@ -30,7 +30,7 @@ class _Session:
 
 
 def test_allowed_models_legacy_empty_list_remains_unrestricted(monkeypatch):
-    monkeypatch.setattr("routes.chat_helpers.get_current_user", lambda request: "alice")
+    monkeypatch.setattr("routes.chat_helpers.effective_user", lambda request: "alice")
 
     _enforce_chat_privileges(
         _Request({"allowed_models": [], "max_messages_per_day": 0}),
@@ -39,7 +39,7 @@ def test_allowed_models_legacy_empty_list_remains_unrestricted(monkeypatch):
 
 
 def test_allowed_models_explicit_empty_restricted_list_blocks_all_models(monkeypatch):
-    monkeypatch.setattr("routes.chat_helpers.get_current_user", lambda request: "alice")
+    monkeypatch.setattr("routes.chat_helpers.effective_user", lambda request: "alice")
 
     with pytest.raises(HTTPException) as exc:
         _enforce_chat_privileges(
@@ -56,7 +56,7 @@ def test_allowed_models_explicit_empty_restricted_list_blocks_all_models(monkeyp
 
 
 def test_allowed_models_nonempty_list_still_restricts_without_new_flag(monkeypatch):
-    monkeypatch.setattr("routes.chat_helpers.get_current_user", lambda request: "alice")
+    monkeypatch.setattr("routes.chat_helpers.effective_user", lambda request: "alice")
 
     _enforce_chat_privileges(
         _Request({"allowed_models": ["provider/model-a"], "max_messages_per_day": 0}),
@@ -70,7 +70,7 @@ def test_allowed_models_nonempty_list_still_restricts_without_new_flag(monkeypat
 
 
 def test_no_restriction_allows_any_model(monkeypatch):
-    monkeypatch.setattr("routes.chat_helpers.get_current_user", lambda request: "alice")
+    monkeypatch.setattr("routes.chat_helpers.effective_user", lambda request: "alice")
 
     privs = {"allowed_models": [], "block_all_models": False, "max_messages_per_day": 0}
     _enforce_chat_privileges(_Request(privs), _Session("provider/model-a"))
@@ -78,7 +78,7 @@ def test_no_restriction_allows_any_model(monkeypatch):
 
 
 def test_specific_allowlist_blocks_models_outside_it(monkeypatch):
-    monkeypatch.setattr("routes.chat_helpers.get_current_user", lambda request: "alice")
+    monkeypatch.setattr("routes.chat_helpers.effective_user", lambda request: "alice")
 
     privs = {
         "allowed_models": ["gpt-4"],
@@ -92,7 +92,7 @@ def test_specific_allowlist_blocks_models_outside_it(monkeypatch):
 
 
 def test_block_all_models_blocks_regardless_of_allowed_models_contents(monkeypatch):
-    monkeypatch.setattr("routes.chat_helpers.get_current_user", lambda request: "alice")
+    monkeypatch.setattr("routes.chat_helpers.effective_user", lambda request: "alice")
 
     # Even if allowed_models contains entries, block_all_models wins.
     privs = {
@@ -111,7 +111,7 @@ def test_block_all_models_blocks_regardless_of_allowed_models_contents(monkeypat
 def test_admin_user_is_never_blocked(monkeypatch):
     from core.auth import ADMIN_PRIVILEGES
 
-    monkeypatch.setattr("routes.chat_helpers.get_current_user", lambda request: "admin")
+    monkeypatch.setattr("routes.chat_helpers.effective_user", lambda request: "admin")
 
     class _AdminAuthManager:
         def get_privileges(self, username):

tests/test_checkin_digest_owner_scope.py

@@ -0,0 +1,70 @@
+"""Check-in calendar digest must be scoped to the task owner.
+
+The digest query selected CalendarEvent with no owner scope, so a scheduled
+check-in for one user pulled EVERY user's calendar events (summaries,
+locations) into their digest — a cross-tenant leak. Ownership lives on
+CalendarCal.owner; the query must join it, like routes/calendar_routes.
+"""
+import tempfile
+import uuid
+from datetime import datetime
+
+import pytest
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+from sqlalchemy.pool import NullPool
+
+import core.database as cdb
+from core.database import CalendarEvent, CalendarCal
+from src.task_scheduler import _checkin_calendar_events
+
+_TMPDB = tempfile.NamedTemporaryFile(suffix=".db", delete=False)
+_ENGINE = create_engine(f"sqlite:///{_TMPDB.name}", connect_args={"check_same_thread": False}, poolclass=NullPool)
+cdb.Base.metadata.create_all(_ENGINE)
+_TS = sessionmaker(bind=_ENGINE, autoflush=False, autocommit=False)
+
+
+def _seed():
+    db = _TS()
+    try:
+        db.query(CalendarEvent).delete(); db.query(CalendarCal).delete()
+        db.add(CalendarCal(id="calA", owner="alice", name="A"))
+        db.add(CalendarCal(id="calB", owner="bob", name="B"))
+        db.add(CalendarEvent(uid="a1", calendar_id="calA", summary="Alice mtg",
+                             dtstart=datetime(2026, 6, 10, 9, 0),
+                             dtend=datetime(2026, 6, 10, 10, 0), status="confirmed"))
+        db.add(CalendarEvent(uid="b1", calendar_id="calB", summary="Bob secret",
+                             dtstart=datetime(2026, 6, 10, 10, 0),
+                             dtend=datetime(2026, 6, 10, 11, 0), status="confirmed"))
+        db.commit()
+    finally:
+        db.close()
+
+
+def test_digest_only_returns_owner_events():
+    _seed()
+    db = _TS()
+    try:
+        s, e = datetime(2026, 6, 1), datetime(2026, 6, 30)
+        alice = _checkin_calendar_events(db, "alice", s, e)
+        assert [ev.summary for ev in alice] == ["Alice mtg"]  # not Bob's
+        bob = _checkin_calendar_events(db, "bob", s, e)
+        assert [ev.summary for ev in bob] == ["Bob secret"]
+    finally:
+        db.close()
+
+
+def test_cancelled_excluded_and_window_respected():
+    _seed()
+    db = _TS()
+    try:
+        db2 = _TS()
+        db2.add(CalendarEvent(uid="a2", calendar_id="calA", summary="cancelled",
+                              dtstart=datetime(2026, 6, 11),
+                              dtend=datetime(2026, 6, 11, 1, 0), status="cancelled"))
+        db2.commit(); db2.close()
+        s, e = datetime(2026, 6, 1), datetime(2026, 6, 30)
+        out = _checkin_calendar_events(db, "alice", s, e)
+        assert "cancelled" not in [ev.summary for ev in out]
+    finally:
+        db.close()

tests/test_companion_readonly.py

@@ -13,6 +13,9 @@ import json
 from types import SimpleNamespace
 from unittest.mock import MagicMock
 
+import pytest
+from fastapi import HTTPException
+
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 # core.database instantiates SQLAlchemy declarative classes at import time, which
@@ -225,12 +228,34 @@ def test_models_route_scopes_api_token_to_token_owner(monkeypatch):
     endpoints = _call_models_route(
         monkeypatch,
         rows,
-        _request(api_token=True, api_token_owner="alice", current_user="api"),
+        _request(
+            api_token=True,
+            api_token_owner="alice",
+            api_token_scopes=["chat"],
+            current_user="api",
+        ),
     )
 
     assert _endpoint_names(endpoints) == ["alice-endpoint", "shared-endpoint"]
 
 
+def test_models_route_rejects_api_token_without_chat_scope(monkeypatch):
+    monkeypatch.setattr(companion_routes, "get_current_user", lambda request: "api")
+
+    with pytest.raises(HTTPException) as exc:
+        _models_route()(
+            _request(
+                api_token=True,
+                api_token_owner="alice",
+                api_token_scopes=["todos:read"],
+                current_user="api",
+            )
+        )
+
+    assert exc.value.status_code == 403
+    assert "chat scope" in exc.value.detail
+
+
 def test_models_route_unresolved_owner_returns_only_shared_rows(monkeypatch):
     rows = [
         _ep(1, "alice-endpoint", "alice"),
@@ -242,7 +267,12 @@ def test_models_route_unresolved_owner_returns_only_shared_rows(monkeypatch):
     endpoints = _call_models_route(
         monkeypatch,
         rows,
-        _request(api_token=True, api_token_owner=None, current_user="api"),
+        _request(
+            api_token=True,
+            api_token_owner=None,
+            api_token_scopes=["chat"],
+            current_user="api",
+        ),
     )
 
     assert _endpoint_names(endpoints) == ["shared-endpoint"]

tests/test_cookbook_helpers.py

@@ -786,6 +786,50 @@ def test_cached_model_scan_reports_plain_dir_gguf(tmp_path):
     assert ggufs[3]["quant"] == "BF16"
 
 
+def test_cached_model_scan_uses_ollama_api_before_cli_and_windows_opt_in():
+    script = _cached_model_scan_script()
+
+    assert "scan_ollama_api()\nscan_ollama()" in script
+    assert "if any(m.get('is_ollama') for m in models): return" in script
+    assert "os.name == 'nt'" in script
+    assert "ODYSSEUS_ALLOW_OLLAMA_CLI_SCAN" in script
+
+
+@pytest.mark.skipif(os.name != "nt", reason="Windows Ollama CLI startup guard")
+def test_cached_model_scan_does_not_launch_ollama_cli_on_windows(tmp_path):
+    """Official Ollama for Windows can auto-start the tray/server on `ollama list`.
+    The read-only cache scanner must not invoke that CLI unless explicitly opted in.
+    """
+    marker = tmp_path / "ollama-called.txt"
+    fake_ollama = tmp_path / "ollama.cmd"
+    fake_ollama.write_text(
+        "@echo off\r\n"
+        f'echo called>"{marker}"\r\n'
+        "echo NAME ID SIZE MODIFIED\r\n"
+        "echo local-model:latest abc 1 GB now\r\n",
+        encoding="utf-8",
+    )
+
+    empty_home = tmp_path / "home"
+    empty_home.mkdir()
+    scan_py = tmp_path / "scan_cache.py"
+    scan_py.write_text(_cached_model_scan_script(), encoding="utf-8")
+    env = dict(os.environ)
+    env["PATH"] = str(tmp_path) + os.pathsep + env.get("PATH", "")
+    env["HOME"] = str(empty_home)
+    env.pop("ODYSSEUS_ALLOW_OLLAMA_CLI_SCAN", None)
+    proc = subprocess.run(
+        [sys.executable, str(scan_py)],
+        check=True,
+        capture_output=True,
+        text=True,
+        env=env,
+    )
+
+    assert marker.exists() is False
+    assert all(m.get("backend") != "ollama" for m in json.loads(proc.stdout))
+
+
 def test_cached_model_scan_uses_huggingface_cache_env(tmp_path):
     """Docker recreates can leave the persisted HF cache outside HOME.
     The Serve scanner should honor the cache env path instead of only ~/.cache.

tests/test_fastembed_cache_path.py

@@ -0,0 +1,69 @@
+"""Regression: FASTEMBED_CACHE_DIR must tolerate a PRESENT-but-EMPTY
+FASTEMBED_CACHE_PATH.
+
+docker-compose.yml injects ``FASTEMBED_CACHE_PATH=${FASTEMBED_CACHE_PATH:-}``,
+which sets the variable to ``""`` when the host has not defined it. The old
+``os.getenv("FASTEMBED_CACHE_PATH", default)`` only used the default when the
+variable was ABSENT, so an empty value made ``FASTEMBED_CACHE_DIR == ""`` →
+``os.makedirs("")`` raised ``[Errno 2] No such file or directory: ''`` →
+FastEmbed failed to initialise and every vector feature (RAG, semantic memory,
+tool index) silently degraded on the default Docker stack.
+
+These tests pin the fix: empty is treated like absent → use the DATA_DIR
+default, while an explicit non-empty override is still honoured.
+"""
+
+from __future__ import annotations
+
+import importlib
+import os
+
+import src.constants as constants
+
+
+def _reload_with(monkeypatch, value):
+    """Reload src.constants with FASTEMBED_CACHE_PATH set to ``value`` (or
+    removed when ``value`` is None) and return the reloaded module."""
+    if value is None:
+        monkeypatch.delenv("FASTEMBED_CACHE_PATH", raising=False)
+    else:
+        monkeypatch.setenv("FASTEMBED_CACHE_PATH", value)
+    return importlib.reload(constants)
+
+
+def _restore(monkeypatch):
+    """Return the module to its env-default state so reloading it here does
+    not leak a test-specific FASTEMBED_CACHE_DIR into other tests."""
+    monkeypatch.delenv("FASTEMBED_CACHE_PATH", raising=False)
+    importlib.reload(constants)
+
+
+def test_empty_fastembed_cache_path_falls_back_to_default(monkeypatch):
+    """The bug: an empty FASTEMBED_CACHE_PATH (exactly what Docker injects)
+    must fall back to the DATA_DIR default, never the empty string."""
+    try:
+        mod = _reload_with(monkeypatch, "")
+        assert mod.FASTEMBED_CACHE_DIR, "empty env must not yield an empty path"
+        assert mod.FASTEMBED_CACHE_DIR == os.path.join(mod.DATA_DIR, "fastembed_cache")
+    finally:
+        _restore(monkeypatch)
+
+
+def test_unset_fastembed_cache_path_uses_default(monkeypatch):
+    """Sanity: an absent variable also resolves to the default."""
+    try:
+        mod = _reload_with(monkeypatch, None)
+        assert mod.FASTEMBED_CACHE_DIR == os.path.join(mod.DATA_DIR, "fastembed_cache")
+    finally:
+        _restore(monkeypatch)
+
+
+def test_explicit_fastembed_cache_path_is_respected(monkeypatch):
+    """A real explicit override must still win — the fix only changes the
+    empty-value handling, not the documented FASTEMBED_CACHE_PATH override."""
+    custom = os.path.join("custom", "fastembed-cache")
+    try:
+        mod = _reload_with(monkeypatch, custom)
+        assert mod.FASTEMBED_CACHE_DIR == custom
+    finally:
+        _restore(monkeypatch)

tests/test_kv_cache_invalidation_2927.py

@@ -79,7 +79,7 @@ def _build_context_harness(monkeypatch, chat_helpers, history):
     monkeypatch.setattr(chat_helpers, "extract_preset", fake_extract_preset)
     monkeypatch.setattr(chat_helpers, "add_user_message", fake_add_user_message)
     monkeypatch.setattr(chat_helpers, "load_prefs_for_user", lambda user: {})
-    monkeypatch.setattr(chat_helpers, "get_current_user", lambda request: "tester")
+    monkeypatch.setattr(chat_helpers, "effective_user", lambda request: "tester")
     monkeypatch.setattr(chat_helpers, "normalize_model_id", lambda endpoint_url, model, **kwargs: None)
     monkeypatch.setattr(chat_helpers, "maybe_compact", fake_maybe_compact)
     monkeypatch.setattr(chat_helpers, "trim_for_context", lambda messages, context_length: messages)

tests/test_mcp_memory_owner_scope.py

@@ -0,0 +1,150 @@
+import asyncio
+
+import mcp_servers.memory_server as memory_server
+from src.memory import MemoryManager
+
+
+class FakeVector:
+    healthy = True
+
+    def __init__(self):
+        self.added = []
+        self.removed = []
+
+    def add(self, memory_id, text):
+        self.added.append((memory_id, text))
+
+    def remove(self, memory_id):
+        self.removed.append(memory_id)
+
+
+def _tool_text(arguments):
+    result = asyncio.run(memory_server.call_tool("manage_memory", arguments))
+    return result[0].text
+
+
+def _entry(manager, text, owner=None, memory_id=None, category="fact"):
+    entry = manager.add_entry(text, owner=owner, category=category)
+    if memory_id:
+        entry["id"] = memory_id
+    return entry
+
+
+def _configure_server(monkeypatch, manager, vector=None):
+    monkeypatch.setattr(memory_server, "_memory_manager", manager)
+    monkeypatch.setattr(memory_server, "_memory_vector", vector)
+    monkeypatch.setattr(memory_server, "_initialized", True)
+    for key in memory_server._OWNER_ENV_KEYS:
+        monkeypatch.delenv(key, raising=False)
+
+
+def test_mcp_memory_uses_configured_owner_for_all_operations(monkeypatch, tmp_path):
+    manager = MemoryManager(str(tmp_path))
+    vector = FakeVector()
+    alice = _entry(
+        manager,
+        "Alice likes green tea",
+        owner="alice",
+        memory_id="aaaaaaaa-0000-0000-0000-000000000000",
+    )
+    bob = _entry(
+        manager,
+        "Bob likes espresso",
+        owner="bob",
+        memory_id="bbbbbbbb-0000-0000-0000-000000000000",
+    )
+    manager.save([alice, bob])
+    _configure_server(monkeypatch, manager, vector)
+    monkeypatch.setenv("ODYSSEUS_MCP_MEMORY_OWNER", "alice")
+
+    list_text = _tool_text({"action": "list"})
+    assert "Alice likes green tea" in list_text
+    assert "Bob likes espresso" not in list_text
+
+    search_text = _tool_text({"action": "search", "text": "likes"})
+    assert "Alice likes green tea" in search_text
+    assert "Bob likes espresso" not in search_text
+
+    add_text = _tool_text({
+        "action": "add",
+        "text": "Alice prefers concise notes",
+        "category": "preference",
+    })
+    assert "Memory added" in add_text
+    added = next(
+        entry for entry in manager.load_all()
+        if entry["text"] == "Alice prefers concise notes"
+    )
+    assert added["owner"] == "alice"
+    assert vector.added == [(added["id"], "Alice prefers concise notes")]
+
+    edit_text = _tool_text({
+        "action": "edit",
+        "memory_id": bob["id"][:8],
+        "text": "Bob changed",
+    })
+    assert edit_text == "Error: Memory 'bbbbbbbb' not found"
+    bob_after_edit = next(
+        entry for entry in manager.load_all()
+        if entry["id"] == bob["id"]
+    )
+    assert bob_after_edit["text"] == "Bob likes espresso"
+
+    delete_text = _tool_text({"action": "delete", "memory_id": bob["id"][:8]})
+    assert delete_text == "Error: Memory 'bbbbbbbb' not found"
+    assert any(entry["id"] == bob["id"] for entry in manager.load_all())
+
+
+def test_mcp_memory_fails_closed_without_owner_for_owner_scoped_store(monkeypatch, tmp_path):
+    manager = MemoryManager(str(tmp_path))
+    alice = _entry(manager, "Alice private memory", owner="alice", memory_id="aaaaaaaa-0000")
+    bob = _entry(manager, "Bob private memory", owner="bob", memory_id="bbbbbbbb-0000")
+    manager.save([alice, bob])
+    _configure_server(monkeypatch, manager, FakeVector())
+    before = manager.load_all()
+
+    actions = [
+        {"action": "list"},
+        {"action": "search", "text": "private"},
+        {"action": "add", "text": "new ownerless memory"},
+        {"action": "edit", "memory_id": alice["id"][:8], "text": "changed"},
+        {"action": "delete", "memory_id": alice["id"][:8]},
+    ]
+
+    for arguments in actions:
+        assert _tool_text(arguments).startswith("Error: Memory MCP owner is not configured")
+
+    assert manager.load_all() == before
+
+
+def test_mcp_memory_preserves_ownerless_local_behavior(monkeypatch, tmp_path):
+    manager = MemoryManager(str(tmp_path))
+    legacy = _entry(
+        manager,
+        "Legacy local memory",
+        memory_id="llllllll-0000-0000-0000-000000000000",
+    )
+    manager.save([legacy])
+    _configure_server(monkeypatch, manager, FakeVector())
+
+    assert "Legacy local memory" in _tool_text({"action": "list"})
+    assert "Legacy local memory" in _tool_text({"action": "search", "text": "legacy"})
+
+    add_text = _tool_text({"action": "add", "text": "Another local memory"})
+    assert "Memory added" in add_text
+    added = next(
+        entry for entry in manager.load_all()
+        if entry["text"] == "Another local memory"
+    )
+    assert "owner" not in added
+
+    assert _tool_text({
+        "action": "edit",
+        "memory_id": legacy["id"][:8],
+        "text": "Updated local memory",
+    }) == "Memory updated: Updated local memory"
+    assert any(entry["text"] == "Updated local memory" for entry in manager.load_all())
+
+    delete_text = _tool_text({"action": "delete", "memory_id": legacy["id"][:8]})
+    assert delete_text.startswith("Memory deleted:")
+    assert all(entry["id"] != legacy["id"] for entry in manager.load_all())

tests/test_review_regressions.py

@@ -385,7 +385,7 @@ async def test_build_chat_context_incognito_does_not_duplicate_current_user_mess
     monkeypatch.setattr(chat_helpers, "extract_preset", fake_extract_preset)
     monkeypatch.setattr(chat_helpers, "add_user_message", fake_add_user_message)
     monkeypatch.setattr(chat_helpers, "load_prefs_for_user", lambda user: {})
-    monkeypatch.setattr(chat_helpers, "get_current_user", lambda request: "tester")
+    monkeypatch.setattr(chat_helpers, "effective_user", lambda request: "tester")
     monkeypatch.setattr(chat_helpers, "normalize_model_id", lambda endpoint_url, model, **kwargs: None)
     monkeypatch.setattr(chat_helpers, "maybe_compact", fake_maybe_compact)
     monkeypatch.setattr(chat_helpers, "trim_for_context", lambda messages, context_length: messages)

tests/test_session_endpoint_owner_scope.py

@@ -52,6 +52,6 @@ def test_chat_endpoint_recovery_paths_are_owner_scoped():
     assert "def _clear_orphaned_session_endpoint(sess, owner:" in chat_routes
     assert "def _recover_empty_session_model(sess, session_id: str, owner:" in chat_routes
     assert "q = owner_filter(q, ModelEndpoint, owner)" in chat_routes
-    assert "resolve_session_auth(sess, session, owner=get_current_user(request))" in chat_routes
+    assert "resolve_session_auth(sess, session, owner=effective_user(request))" in chat_routes
     assert "def resolve_session_auth(sess, session_id: str, owner:" in chat_helpers
     assert "update_q = update_q.filter(DBSession.owner == owner)" in chat_helpers

tests/test_web_user_agent_constant.py

@@ -0,0 +1,18 @@
+"""The web scraping path routes its User-Agent through one constant.
+
+Guards the dedup: web_fetch / web_search outbound UAs go through
+WEB_FETCH_USER_AGENT, so a stale or bare Mozilla string cannot be re-inlined in
+the search sources.
+"""
+from pathlib import Path
+
+_SEARCH = Path(__file__).resolve().parent.parent / "services" / "search"
+
+
+def test_search_sources_have_no_inline_mozilla_ua():
+    offenders = [
+        str(py.relative_to(_SEARCH.parent.parent))
+        for py in _SEARCH.rglob("*.py")
+        if "Mozilla/" in py.read_text(encoding="utf-8")
+    ]
+    assert not offenders, f"inline Mozilla UA found; use WEB_FETCH_USER_AGENT: {offenders}"

tests/test_webhook_emitters_use_manager.py

@@ -0,0 +1,54 @@
+"""Guard: every public webhook emitter goes through the manager.
+
+Public emitters in `routes/` must schedule their fire through
+`webhook_manager.fire_and_forget(...)` (or `_spawn_tracked`). A bare
+`asyncio.create_task(webhook_manager.fire(...))` escapes
+`WebhookManager._bg_tasks`, so asyncio only holds a weak reference to the
+delivery task and the GC can collect it before it sends — silently dropping
+the webhook. Catching this with a scan stops a regression from sneaking
+back in via a copy-paste.
+"""
+import ast
+from pathlib import Path
+
+ROUTES_DIR = Path(__file__).resolve().parent.parent / "routes"
+
+
+def _untracked_fire_calls(tree: ast.AST) -> list[tuple[int, str]]:
+    """Return (lineno, snippet) for any asyncio.create_task(webhook_manager.fire(...))."""
+    hits: list[tuple[int, str]] = []
+    for node in ast.walk(tree):
+        if not isinstance(node, ast.Call):
+            continue
+        func = node.func
+        if not (isinstance(func, ast.Attribute) and func.attr == "create_task"):
+            continue
+        if not (isinstance(func.value, ast.Name) and func.value.id == "asyncio"):
+            continue
+        if not node.args:
+            continue
+        inner = node.args[0]
+        if not isinstance(inner, ast.Call):
+            continue
+        inner_func = inner.func
+        if (
+            isinstance(inner_func, ast.Attribute)
+            and inner_func.attr == "fire"
+            and isinstance(inner_func.value, ast.Name)
+            and inner_func.value.id == "webhook_manager"
+        ):
+            hits.append((node.lineno, ast.unparse(node)))
+    return hits
+
+
+def test_no_untracked_webhook_fire_in_routes():
+    offenders: list[str] = []
+    for path in ROUTES_DIR.rglob("*.py"):
+        tree = ast.parse(path.read_text(), filename=str(path))
+        for lineno, snippet in _untracked_fire_calls(tree):
+            offenders.append(f"{path.relative_to(ROUTES_DIR.parent)}:{lineno}: {snippet}")
+    assert not offenders, (
+        "Public webhook emitters must use webhook_manager.fire_and_forget(...) "
+        "so the delivery task is tracked in WebhookManager._bg_tasks. Found "
+        "untracked emitter(s):\n  " + "\n  ".join(offenders)
+    )