Add Matrix homeserver with MatrixRTC calling support

- Synapse + PostgreSQL + coturn + LiveKit + lk-jwt-service - Caddy entries for mtx.alogins.net, lk.alogins.net, lkjwt.alogins.net - well-known endpoints for Matrix client/server discovery and RTC transport - Users: admin, elizaveta, aleksandra Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add users backup script with Zabbix notification
2026-03-15 14:12:13 +00:00 · 2026-03-09 06:26:32 +00:00 · 2026-03-08 16:12:13 +00:00 · 2026-03-08 16:06:14 +00:00 · 2026-03-08 15:45:08 +00:00 · 2026-03-08 15:36:30 +00:00
33 changed files with 1110 additions and 1755 deletions
--- a/.gitignore
+++ b/.gitignore
@@ -0,0 +1,2 @@
+adolf/.env
+seafile/.env
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -13,6 +13,7 @@ This repository manages Docker Compose configurations for the **Agap** self-host
 | `immich-app/` | Immich (photo management) | 2283 | Main compose via root `docker-compose.yml` |
 | `gitea/` | Gitea (git hosting) + Postgres | 3000, 222 | Standalone compose |
 | `openai/` | Open WebUI + Ollama (AI chat) | 3125 | Requires NVIDIA GPU |
+| `vaultwarden/` | Vaultwarden (password manager) | 8041 | Backup script in `vaultwarden/backup.sh` |

 ## Common Commands

@@ -90,6 +91,8 @@ When changes are made to infrastructure (services, config, setup), update the re
 | Home-Assistant | KVM-based Home Assistant setup |
 | 3X-UI | VPN proxy panel |
 | Gitea | Git hosting Docker service |
+| Vaultwarden | Password manager, CLI setup, backup |
+| Seafile | File sync, document editing, OnlyOffice, WebDAV |

 ### Read Wiki Pages (API)

@@ -125,3 +128,100 @@ git push http://alvis:$GITEA_TOKEN@localhost:3000/alvis/AgapHost.wiki.git main
 - Remove outdated or redundant content when updating
 - Create a new page if a topic doesn't exist yet
 - Wiki files are Markdown, named `<PageTitle>.md`
+
+## Home Assistant API
+
+**Instance**: `https://haos.alogins.net`
+**Token**: Read from `$HA_TOKEN` environment variable — never hardcode it
+**Base URL**: `https://haos.alogins.net/api/`
+**Auth header**: `Authorization: Bearer <token>`
+
+### Common Endpoints
+```bash
+# Health check
+curl -s -H "Authorization: Bearer $HA_TOKEN" \
+  https://haos.alogins.net/api/
+
+# Get all entity states
+curl -s -H "Authorization: Bearer $HA_TOKEN" \
+  https://haos.alogins.net/api/states
+
+# Get specific entity
+curl -s -H "Authorization: Bearer $HA_TOKEN" \
+  https://haos.alogins.net/api/states/<entity_id>
+
+# Call service (e.g., turn on light)
+curl -s -X POST \
+  -H "Authorization: Bearer $HA_TOKEN" \
+  -H "Content-Type: application/json" \
+  -d '{"entity_id":"light.example"}' \
+  https://haos.alogins.net/api/services/<domain>/<service>
+```
+
+**Note**: Status 401 = token invalid/expired
+
+## HA → Zabbix Alerting
+
+Home Assistant automations push alerts to Zabbix via `history.push` API (Zabbix 7.4 trapper items). No middleware needed.
+
+### Architecture
+
+```
+[HA sensor ON] → [HA automation] → [rest_command: HTTP POST] → [Zabbix history.push] → [trapper item] → [trigger] → [Telegram]
+```
+
+### Water Leak Sensors
+
+3x HOBEIAN ZG-222Z moisture sensors → Disaster-level Zabbix alert with room name.
+
+| HA Entity | Room |
+|-----------|------|
+| `binary_sensor.hobeian_zg_222z` | Kitchen |
+| `binary_sensor.hobeian_zg_222z_2` | Bathroom |
+| `binary_sensor.hobeian_zg_222z_3` | Laundry |
+
+**Zabbix side** (host "HA Agap", hostid 10780):
+- Trapper item: `water.leak` (text type) — receives room name or "ok"
+- Trigger: `last(/HA Agap/water.leak)<>"ok"` — Disaster (severity 5), manual close
+- Trigger name uses `{ITEM.LASTVALUE}` to show room in notification
+
+**HA side** (`configuration.yaml`):
+- `rest_command.zabbix_water_leak` — POST to Zabbix `history.push`, accepts `{{ room }}` template variable
+- `rest_command.zabbix_water_leak_clear` — pushes "ok" to clear
+- Automation "Water Leak Alert" — any sensor ON → sends room name to Zabbix
+- Automation "Water Leak Clear" — all sensors OFF → sends "ok"
+
+### Adding a New HA → Zabbix Alert
+
+1. **Zabbix**: Create trapper item (type 2) on "HA Agap" via `item.create` API. Create trigger via `trigger.create`.
+2. **HA config**: Add `rest_command` entry in `configuration.yaml` with `history.push` payload. Restart HA.
+3. **HA automation**: Create via `POST /api/config/automation/config/<id>` with trigger on sensor state and action calling the rest_command.
+4. **Test**: Call `rest_command` via HA API, verify Zabbix problem appears.
+
+## Zabbix API
+
+**Instance**: `http://localhost:81` (local), `https://zb.alogins.net` (external)
+**Endpoint**: `http://localhost:81/api_jsonrpc.php`
+**Token**: Read from `$ZABBIX_TOKEN` environment variable — never hardcode it
+**Auth header**: `Authorization: Bearer <token>`
+
+### Common Requests
+```bash
+# Check API version
+curl -s -X POST http://localhost:81/api_jsonrpc.php \
+  -H "Content-Type: application/json" \
+  -H "Authorization: Bearer $ZABBIX_TOKEN" \
+  -d '{"jsonrpc":"2.0","method":"apiinfo.version","params":{},"id":1}'
+
+# Get all hosts
+curl -s -X POST http://localhost:81/api_jsonrpc.php \
+  -H "Content-Type: application/json" \
+  -H "Authorization: Bearer $ZABBIX_TOKEN" \
+  -d '{"jsonrpc":"2.0","method":"host.get","params":{"output":"extend"},"id":1}'
+
+# Get problems/issues
+curl -s -X POST http://localhost:81/api_jsonrpc.php \
+  -H "Content-Type: application/json" \
+  -H "Authorization: Bearer $ZABBIX_TOKEN" \
+  -d '{"jsonrpc":"2.0","method":"problem.get","params":{"output":"extend"},"id":1}'
+```
--- a/122
+++ b/122
@@ -0,0 +1,122 @@
+haos.alogins.net {
+	reverse_proxy http://192.168.1.141:8123 {
+	
+        header_up X-Forwarded-For {remote_host}
+        header_up X-Forwarded-Proto {scheme}
+}
+}
+
+
+vi.alogins.net {
+	reverse_proxy localhost:2283
+}
+
+doc.alogins.net {
+	reverse_proxy localhost:11001
+}
+
+zb.alogins.net {
+	reverse_proxy localhost:81
+}
+
+wiki.alogins.net {
+	reverse_proxy localhost:8083 {
+        	header_up Host {http.request.host}
+        	header_up X-Forwarded-Proto {scheme}
+        	header_up X-Real-IP {remote_host}
+	}
+}
+
+nn.alogins.net {
+	reverse_proxy localhost:5678
+}
+
+git.alogins.net {
+	reverse_proxy localhost:3000
+}
+
+ds.alogins.net {
+	reverse_proxy localhost:3974
+}
+
+ai.alogins.net {
+	reverse_proxy localhost:3125
+}
+
+openpi.alogins.net {
+	root * /home/alvis/tmp/files/pi05_droid
+    	file_server browse
+
+}
+
+
+vui3.alogins.net {
+    @xhttp {
+        path /VLSpdG9k/xht*
+    }
+    handle @xhttp {
+        reverse_proxy http://localhost:8445 {
+            flush_interval -1
+            header_up X-Real-IP {remote_host}
+	    transport http {
+              read_timeout 0
+              write_timeout 0
+              dial_timeout 10s
+            }
+        }
+    }
+	reverse_proxy /gnYCNq4EbYukS5qtOe/* localhost:58959
+	respond 401 
+}
+
+vui4.alogins.net {
+	reverse_proxy localhost:58959
+}
+
+ntfy.alogins.net {
+	reverse_proxy localhost:8840
+}
+
+docs.alogins.net {
+	reverse_proxy localhost:8078
+}
+
+office.alogins.net {
+	reverse_proxy localhost:6233
+}
+
+vw.alogins.net {
+	reverse_proxy localhost:8041
+}
+
+mtx.alogins.net {
+	handle /.well-known/matrix/client {
+		header Content-Type application/json
+		header Access-Control-Allow-Origin *
+		respond `{"m.homeserver":{"base_url":"https://mtx.alogins.net"},"org.matrix.msc4143.rtc_foci":[{"type":"livekit","livekit_service_url":"https://lkjwt.alogins.net"}]}`
+	}
+	handle /.well-known/matrix/server {
+		header Content-Type application/json
+		header Access-Control-Allow-Origin *
+		respond `{"m.server":"mtx.alogins.net:443"}`
+	}
+	handle /_matrix/client/unstable/org.matrix.msc4143/rtc/transports {
+		header Content-Type application/json
+		header Access-Control-Allow-Origin *
+		respond `{"foci":[{"type":"livekit","livekit_service_url":"https://lkjwt.alogins.net"}]}`
+	}
+	reverse_proxy localhost:8008
+}
+
+lkjwt.alogins.net {
+	reverse_proxy localhost:8009
+}
+
+lk.alogins.net {
+	reverse_proxy localhost:7880
+}
+
+localhost:8042 {
+	reverse_proxy localhost:8041
+    	tls internal
+}
--- a/adolf/ARCHITECTURE.md
+++ b/adolf/ARCHITECTURE.md
@@ -1,144 +0,0 @@
-# Adolf
-
-Persistent AI assistant reachable via Telegram. Three-tier model routing with GPU VRAM management.
-
-## Architecture
-
-```
-Telegram user
-     ↕ (long-polling)
-[grammy] Node.js — port 3001
-  - grammY bot polls Telegram
-  - on message: fire-and-forget POST /chat to deepagents
-  - exposes MCP SSE server: tool send_telegram_message(chat_id, text)
-     ↓ POST /chat → 202 Accepted immediately
-[deepagents] Python FastAPI — port 8000
-     ↓
-Pre-check: starts with /think? → force_complex=True, strip prefix
-     ↓
-Router (qwen2.5:0.5b, ~1-2s, always warm in VRAM)
-  Structured output: {tier: light|medium|complex, confidence: 0.0-1.0, reply?: str}
-  - light:   simple conversational → router answers directly, ~1-2s
-  - medium:  needs memory/web search → qwen3:4b + deepagents tools
-  - complex: multi-step research, planning, code → qwen3:8b + subagents
-  force_complex always overrides to complex
-  complex only if confidence >= 0.85 (else downgraded to medium)
-     ↓
-     ├── light ─────────── router reply used directly (no extra LLM call)
-     ├── medium ──────────  deepagents qwen3:4b + TodoList + tools
-     └── complex ─────────  VRAM flush → deepagents qwen3:8b + TodoList + subagents
-                             └→ background: exit_complex_mode (flush 8b, prewarm 4b+router)
-     ↓
-send_telegram_message via grammy MCP
-     ↓
-asyncio.create_task(store_memory_async) — spin-wait GPU idle → openmemory add_memory
-     ↕ MCP SSE                         ↕ HTTP
-[openmemory] Python + mem0 — port 8765   [SearXNG — port 11437]
-  - add_memory, search_memory, get_all_memories
-  - extractor: qwen2.5:1.5b on GPU Ollama (11436) — 2–5s
-  - embedder: nomic-embed-text on CPU Ollama (11435) — 50–150ms
-  - vector store: Qdrant (port 6333), 768 dims
-```
-
-## Three-Tier Model Routing
-
-| Tier | Model | VRAM | Trigger | Latency |
-|------|-------|------|---------|---------|
-| Light | qwen2.5:1.5b (router answers) | ~1.2 GB (shared with extraction) | Router classifies as light | ~2–4s |
-| Medium | qwen3:4b | ~2.5 GB | Default; router classifies medium | ~20–40s |
-| Complex | qwen3:8b | ~5.5 GB | `/think` prefix | ~60–120s |
-
-**Normal VRAM** (light + medium): router/extraction(1.2, shared) + medium(2.5) = ~3.7 GB
-**Complex VRAM**: 8b alone = ~5.5 GB — must flush others first
-
-### Router model: qwen2.5:1.5b (not 0.5b)
-
-qwen2.5:0.5b is too small for reliable classification — tends to output "medium" for everything
-or produces nonsensical output. qwen2.5:1.5b is already loaded in VRAM for memory extraction,
-so switching adds zero net VRAM overhead while dramatically improving accuracy.
-
-Router uses **raw text generation** (not structured output/JSON schema):
- Ask model to output one word: `light`, `medium`, or `complex`
- Parse with simple keyword matching (fallback: `medium`)
- For `light` tier: a second call generates the reply text
-
-## VRAM Management
-
-GTX 1070 has 8 GB VRAM. Ollama's auto-eviction can spill models to CPU RAM permanently
-(all subsequent loads stay on CPU). To prevent this:
-
-1. **Always flush explicitly** before loading qwen3:8b (`keep_alive=0`)
-2. **Verify eviction** via `/api/ps` poll (15s timeout) before proceeding
-3. **Fallback**: timeout → log warning, run medium agent instead
-4. **Post-complex**: flush 8b immediately, pre-warm 4b + router
-
-```python
-# Flush (force immediate unload):
-POST /api/generate {"model": "qwen3:4b", "prompt": "", "keep_alive": 0}
-
-# Pre-warm (load into VRAM for 5 min):
-POST /api/generate {"model": "qwen3:4b", "prompt": "", "keep_alive": 300}
-```
-
-## Agents
-
-**Medium agent** (`build_medium_agent`):
- `create_deep_agent` with TodoListMiddleware (auto-included)
- Tools: `search_memory`, `get_all_memories`, `web_search`
- No subagents
-
-**Complex agent** (`build_complex_agent`):
- `create_deep_agent` with TodoListMiddleware + SubAgentMiddleware
- Tools: all agent tools
- Subagents:
-  - `research`: web_search only, for thorough multi-query web research
-  - `memory`: search_memory + get_all_memories, for comprehensive context retrieval
-
-## Concurrency
-
-| Semaphore | Guards | Notes |
-|-----------|--------|-------|
-| `_reply_semaphore(1)` | GPU Ollama (all tiers) | One LLM reply inference at a time |
-| `_memory_semaphore(1)` | GPU Ollama (qwen2.5:1.5b extraction) | One memory extraction at a time |
-
-Light path holds `_reply_semaphore` briefly (no GPU inference).
-Memory extraction spin-waits until `_reply_semaphore` is free (60s timeout).
-
-## Pipeline
-
-1. User message → Grammy → `POST /chat` → 202 Accepted
-2. Background: acquire `_reply_semaphore` → route → run agent tier → send reply
-3. `asyncio.create_task(store_memory_async)` — spin-waits GPU free, then extracts memories
-4. For complex: `asyncio.create_task(exit_complex_mode)` — flushes 8b, pre-warms 4b+router
-
-## External Services (from openai/ stack)
-
-| Service | Host Port | Role |
-|---------|-----------|------|
-| Ollama GPU | 11436 | All reply inference + extraction (qwen2.5:1.5b) |
-| Ollama CPU | 11435 | Memory embedding (nomic-embed-text) |
-| Qdrant | 6333 | Vector store for memories |
-| SearXNG | 11437 | Web search |
-
-GPU Ollama config: `OLLAMA_MAX_LOADED_MODELS=2`, `OLLAMA_NUM_PARALLEL=1`.
-
-## Files
-
-```
-adolf/
-├── docker-compose.yml      Services: deepagents, openmemory, grammy
-├── Dockerfile              deepagents container (Python 3.12)
-├── agent.py                FastAPI + three-tier routing + run_agent_task
-├── router.py               Router class — qwen2.5:0.5b structured output routing
-├── vram_manager.py         VRAMManager — flush/prewarm/poll Ollama VRAM
-├── agent_factory.py        build_medium_agent / build_complex_agent (deepagents)
-├── .env                    TELEGRAM_BOT_TOKEN (not committed)
-├── openmemory/
-│   ├── server.py           FastMCP + mem0 MCP tools
-│   ├── requirements.txt
-│   └── Dockerfile
-└── grammy/
-    ├── bot.mjs             grammY bot + MCP SSE server
-    ├── package.json
-    └── Dockerfile
-```
--- a/adolf/Dockerfile
+++ b/adolf/Dockerfile
@@ -1,10 +0,0 @@
-FROM python:3.12-slim
-
-WORKDIR /app
-
-RUN pip install --no-cache-dir deepagents langchain-ollama langgraph \
-    fastapi uvicorn langchain-mcp-adapters langchain-community httpx
-
-COPY agent.py vram_manager.py router.py agent_factory.py hello_world.py .
-
-CMD ["uvicorn", "agent:app", "--host", "0.0.0.0", "--port", "8000"]
--- a/adolf/agent.py
+++ b/adolf/agent.py
@@ -1,309 +0,0 @@
-import asyncio
-import os
-import time
-from contextlib import asynccontextmanager
-
-from fastapi import FastAPI, BackgroundTasks
-from fastapi.responses import JSONResponse
-from pydantic import BaseModel
-
-from langchain_ollama import ChatOllama
-from langchain_mcp_adapters.client import MultiServerMCPClient
-from langchain_community.utilities import SearxSearchWrapper
-from langchain_core.tools import Tool
-
-from vram_manager import VRAMManager
-from router import Router
-from agent_factory import build_medium_agent, build_complex_agent
-
-OLLAMA_BASE_URL = os.getenv("OLLAMA_BASE_URL", "http://localhost:11434")
-ROUTER_MODEL = os.getenv("DEEPAGENTS_ROUTER_MODEL", "qwen2.5:0.5b")
-MEDIUM_MODEL = os.getenv("DEEPAGENTS_MODEL", "qwen3:4b")
-COMPLEX_MODEL = os.getenv("DEEPAGENTS_COMPLEX_MODEL", "qwen3:8b")
-SEARXNG_URL = os.getenv("SEARXNG_URL", "http://host.docker.internal:11437")
-OPENMEMORY_URL = os.getenv("OPENMEMORY_URL", "http://openmemory:8765")
-GRAMMY_URL = os.getenv("GRAMMY_URL", "http://grammy:3001")
-
-MAX_HISTORY_TURNS = 5
-_conversation_buffers: dict[str, list] = {}
-
-MEDIUM_SYSTEM_PROMPT = (
-    "You are a helpful AI assistant talking to a user via Telegram. "
-    "The user's ID is {user_id}. "
-    "IMPORTANT: When calling any memory tool (search_memory, get_all_memories), "
-    "always use user_id=\"{user_id}\". "
-    "Every conversation is automatically saved to memory after you reply — "
-    "you do NOT need to explicitly store anything. "
-    "NEVER tell the user you cannot remember or store information. "
-    "If the user asks you to remember something, acknowledge it and confirm it will be remembered. "
-    "Use search_memory when context from past conversations may be relevant. "
-    "Use web_search for questions about current events or facts you don't know. "
-    "Reply concisely."
-)
-
-COMPLEX_SYSTEM_PROMPT = (
-    "You are a capable AI assistant tackling a complex, multi-step task for a Telegram user. "
-    "The user's ID is {user_id}. "
-    "IMPORTANT: When calling any memory tool (search_memory, get_all_memories), "
-    "always use user_id=\"{user_id}\". "
-    "Plan your work using write_todos before diving in. "
-    "Delegate: use the 'research' subagent for thorough web research across multiple queries, "
-    "and the 'memory' subagent to gather comprehensive context from past conversations. "
-    "Every conversation is automatically saved to memory — you do NOT need to store anything. "
-    "NEVER tell the user you cannot remember or store information. "
-    "Produce a thorough, well-structured reply."
-)
-
-medium_agent = None
-complex_agent = None
-router: Router = None
-vram_manager: VRAMManager = None
-mcp_client = None
-send_tool = None
-add_memory_tool = None
-
-# GPU mutex: one LLM inference at a time
-_reply_semaphore = asyncio.Semaphore(1)
-# Memory semaphore: one async extraction at a time
-_memory_semaphore = asyncio.Semaphore(1)
-
-
-@asynccontextmanager
-async def lifespan(app: FastAPI):
-    global medium_agent, complex_agent, router, vram_manager
-    global mcp_client, send_tool, add_memory_tool
-
-    # Three model instances
-    router_model = ChatOllama(
-        model=ROUTER_MODEL, base_url=OLLAMA_BASE_URL, think=False, num_ctx=4096,
-        temperature=0,  # deterministic classification
-    )
-    medium_model = ChatOllama(
-        model=MEDIUM_MODEL, base_url=OLLAMA_BASE_URL, think=False, num_ctx=8192
-    )
-    complex_model = ChatOllama(
-        model=COMPLEX_MODEL, base_url=OLLAMA_BASE_URL, think=True, num_ctx=16384
-    )
-
-    vram_manager = VRAMManager(base_url=OLLAMA_BASE_URL)
-    router = Router(model=router_model)
-
-    mcp_connections = {
-        "openmemory": {"transport": "sse", "url": f"{OPENMEMORY_URL}/sse"},
-        "grammy": {"transport": "sse", "url": f"{GRAMMY_URL}/sse"},
-    }
-    mcp_client = MultiServerMCPClient(mcp_connections)
-    for attempt in range(12):
-        try:
-            mcp_tools = await mcp_client.get_tools()
-            break
-        except Exception as e:
-            if attempt == 11:
-                raise
-            print(f"[agent] MCP not ready (attempt {attempt + 1}/12): {e}. Retrying in 5s...")
-            await asyncio.sleep(5)
-
-    send_tool = next((t for t in mcp_tools if t.name == "send_telegram_message"), None)
-    add_memory_tool = next((t for t in mcp_tools if t.name == "add_memory"), None)
-    agent_tools = [t for t in mcp_tools if t.name not in ("send_telegram_message", "add_memory")]
-
-    searx = SearxSearchWrapper(searx_host=SEARXNG_URL)
-    agent_tools.append(Tool(
-        name="web_search",
-        func=searx.run,
-        description="Search the web for current information",
-    ))
-
-    # Build agents (system_prompt filled per-request with user_id)
-    medium_agent = build_medium_agent(
-        model=medium_model,
-        agent_tools=agent_tools,
-        system_prompt=MEDIUM_SYSTEM_PROMPT.format(user_id="{user_id}"),
-    )
-    complex_agent = build_complex_agent(
-        model=complex_model,
-        agent_tools=agent_tools,
-        system_prompt=COMPLEX_SYSTEM_PROMPT.format(user_id="{user_id}"),
-    )
-
-    print(
-        f"[agent] three-tier: router={ROUTER_MODEL} | medium={MEDIUM_MODEL} | complex={COMPLEX_MODEL}",
-        flush=True,
-    )
-    print(f"[agent] agent tools: {[t.name for t in agent_tools]}", flush=True)
-
-    yield
-
-    medium_agent = None
-    complex_agent = None
-    router = None
-    vram_manager = None
-    mcp_client = None
-    send_tool = None
-    add_memory_tool = None
-
-
-app = FastAPI(lifespan=lifespan)
-
-
-class ChatRequest(BaseModel):
-    message: str
-    chat_id: str
-
-
-async def store_memory_async(conversation: str, user_id: str):
-    """Fire-and-forget: extract and store memories after GPU is free."""
-    t_wait = time.monotonic()
-    while _reply_semaphore.locked():
-        if time.monotonic() - t_wait > 60:
-            print(f"[memory] spin-wait timeout 60s, proceeding for user {user_id}", flush=True)
-            break
-        await asyncio.sleep(0.5)
-    async with _memory_semaphore:
-        t0 = time.monotonic()
-        try:
-            await add_memory_tool.ainvoke({"text": conversation, "user_id": user_id})
-            print(f"[memory] stored in {time.monotonic() - t0:.1f}s for user {user_id}", flush=True)
-        except Exception as e:
-            print(f"[memory] error after {time.monotonic() - t0:.1f}s: {e}", flush=True)
-
-
-def _extract_final_text(result) -> str | None:
-    """Extract last AIMessage content from agent result."""
-    msgs = result.get("messages", [])
-    for m in reversed(msgs):
-        if type(m).__name__ == "AIMessage" and getattr(m, "content", ""):
-            return m.content
-    # deepagents may return output differently
-    if isinstance(result, dict) and result.get("output"):
-        return result["output"]
-    return None
-
-
-def _log_messages(result):
-    msgs = result.get("messages", [])
-    for m in msgs:
-        role = type(m).__name__
-        content = getattr(m, "content", "")
-        tool_calls = getattr(m, "tool_calls", [])
-        if content:
-            print(f"[agent]   {role}: {str(content)[:150]}", flush=True)
-        for tc in tool_calls:
-            print(f"[agent]   {role} → {tc['name']}({tc['args']})", flush=True)
-
-
-async def run_agent_task(message: str, chat_id: str):
-    print(f"[agent] queued: {message[:80]!r} chat={chat_id}", flush=True)
-
-    # Pre-check: /think prefix forces complex tier
-    force_complex = False
-    clean_message = message
-    if message.startswith("/think "):
-        force_complex = True
-        clean_message = message[len("/think "):]
-        print("[agent] /think prefix → force_complex=True", flush=True)
-
-    async with _reply_semaphore:
-        t0 = time.monotonic()
-        history = _conversation_buffers.get(chat_id, [])
-        print(f"[agent] running: {clean_message[:80]!r}", flush=True)
-
-        # Route the message
-        tier, light_reply = await router.route(clean_message, history, force_complex)
-        print(f"[agent] tier={tier} message={clean_message[:60]!r}", flush=True)
-
-        final_text = None
-        try:
-            if tier == "light":
-                final_text = light_reply
-                llm_elapsed = time.monotonic() - t0
-                print(f"[agent] light path: answered by router", flush=True)
-
-            elif tier == "medium":
-                system_prompt = MEDIUM_SYSTEM_PROMPT.format(user_id=chat_id)
-                result = await medium_agent.ainvoke({
-                    "messages": [
-                        {"role": "system", "content": system_prompt},
-                        *history,
-                        {"role": "user", "content": clean_message},
-                    ]
-                })
-                llm_elapsed = time.monotonic() - t0
-                _log_messages(result)
-                final_text = _extract_final_text(result)
-
-            else:  # complex
-                ok = await vram_manager.enter_complex_mode()
-                if not ok:
-                    print("[agent] complex→medium fallback (eviction timeout)", flush=True)
-                    tier = "medium"
-                    system_prompt = MEDIUM_SYSTEM_PROMPT.format(user_id=chat_id)
-                    result = await medium_agent.ainvoke({
-                        "messages": [
-                            {"role": "system", "content": system_prompt},
-                            *history,
-                            {"role": "user", "content": clean_message},
-                        ]
-                    })
-                else:
-                    system_prompt = COMPLEX_SYSTEM_PROMPT.format(user_id=chat_id)
-                    result = await complex_agent.ainvoke({
-                        "messages": [
-                            {"role": "system", "content": system_prompt},
-                            *history,
-                            {"role": "user", "content": clean_message},
-                        ]
-                    })
-                    asyncio.create_task(vram_manager.exit_complex_mode())
-
-                llm_elapsed = time.monotonic() - t0
-                _log_messages(result)
-                final_text = _extract_final_text(result)
-
-        except Exception as e:
-            import traceback
-            llm_elapsed = time.monotonic() - t0
-            print(f"[agent] error after {llm_elapsed:.1f}s for chat {chat_id}: {e}", flush=True)
-            traceback.print_exc()
-
-        # Send reply via grammy MCP (split if > Telegram's 4096-char limit)
-        if final_text and send_tool:
-            t1 = time.monotonic()
-            MAX_TG = 4000  # leave headroom below the 4096 hard limit
-            chunks = [final_text[i:i + MAX_TG] for i in range(0, len(final_text), MAX_TG)]
-            for chunk in chunks:
-                await send_tool.ainvoke({"chat_id": chat_id, "text": chunk})
-            send_elapsed = time.monotonic() - t1
-            # Log in format compatible with test_pipeline.py parser
-            print(
-                f"[agent] replied in {time.monotonic() - t0:.1f}s "
-                f"(llm={llm_elapsed:.1f}s, send={send_elapsed:.1f}s) tier={tier}",
-                flush=True,
-            )
-        elif not final_text:
-            print("[agent] warning: no text reply from agent", flush=True)
-
-        # Update conversation buffer
-        if final_text:
-            buf = _conversation_buffers.get(chat_id, [])
-            buf.append({"role": "user", "content": clean_message})
-            buf.append({"role": "assistant", "content": final_text})
-            _conversation_buffers[chat_id] = buf[-(MAX_HISTORY_TURNS * 2):]
-
-        # Async memory storage (fire-and-forget)
-        if add_memory_tool and final_text:
-            conversation = f"User: {clean_message}\nAssistant: {final_text}"
-            asyncio.create_task(store_memory_async(conversation, chat_id))
-
-
-@app.post("/chat")
-async def chat(request: ChatRequest, background_tasks: BackgroundTasks):
-    if medium_agent is None:
-        return JSONResponse(status_code=503, content={"error": "Agent not ready"})
-    background_tasks.add_task(run_agent_task, request.message, request.chat_id)
-    return JSONResponse(status_code=202, content={"status": "accepted"})
-
-
-@app.get("/health")
-async def health():
-    return {"status": "ok", "agent_ready": medium_agent is not None}
--- a/adolf/agent_factory.py
+++ b/adolf/agent_factory.py
@@ -1,54 +0,0 @@
-from deepagents import create_deep_agent, SubAgent
-
-
-def build_medium_agent(model, agent_tools: list, system_prompt: str):
-    """Medium agent: create_deep_agent with TodoList planning, no subagents."""
-    return create_deep_agent(
-        model=model,
-        tools=agent_tools,
-        system_prompt=system_prompt,
-    )
-
-
-def build_complex_agent(model, agent_tools: list, system_prompt: str):
-    """Complex agent: create_deep_agent with TodoList planning + research/memory subagents."""
-    web_tools = [t for t in agent_tools if getattr(t, "name", "") == "web_search"]
-    memory_tools = [
-        t for t in agent_tools
-        if getattr(t, "name", "") in ("search_memory", "get_all_memories")
-    ]
-
-    research_sub: SubAgent = {
-        "name": "research",
-        "description": (
-            "Runs multiple web searches in parallel and synthesizes findings. "
-            "Use for thorough research tasks requiring several queries."
-        ),
-        "system_prompt": (
-            "You are a research specialist. Search the web thoroughly using multiple queries. "
-            "Cite sources and synthesize information into a clear summary."
-        ),
-        "tools": web_tools,
-        "model": model,
-    }
-
-    memory_sub: SubAgent = {
-        "name": "memory",
-        "description": (
-            "Searches and retrieves all relevant memories about the user comprehensively. "
-            "Use to gather full context from past conversations."
-        ),
-        "system_prompt": (
-            "You are a memory specialist. Search broadly using multiple queries. "
-            "Return all relevant facts and context you find."
-        ),
-        "tools": memory_tools,
-        "model": model,
-    }
-
-    return create_deep_agent(
-        model=model,
-        tools=agent_tools,
-        system_prompt=system_prompt,
-        subagents=[research_sub, memory_sub],
-    )
--- a/adolf/docker-compose.yml
+++ b/adolf/docker-compose.yml
@@ -1,43 +0,0 @@
-services:
-  deepagents:
-    build: .
-    container_name: deepagents
-    ports:
-      - "8000:8000"
-    environment:
-      - PYTHONUNBUFFERED=1
-      - OLLAMA_BASE_URL=http://host.docker.internal:11436
-      - DEEPAGENTS_MODEL=qwen3:4b
-      - DEEPAGENTS_COMPLEX_MODEL=qwen3:8b
-      - DEEPAGENTS_ROUTER_MODEL=qwen2.5:1.5b
-      - SEARXNG_URL=http://host.docker.internal:11437
-    extra_hosts:
-      - "host.docker.internal:host-gateway"
-    depends_on:
-      - openmemory
-      - grammy
-    restart: unless-stopped
-
-  openmemory:
-    build: ./openmemory
-    container_name: openmemory
-    ports:
-      - "8765:8765"
-    environment:
-      # Extraction LLM (qwen2.5:1.5b) runs on GPU after reply — fast 2-5s extraction
-      - OLLAMA_GPU_URL=http://host.docker.internal:11436
-      # Embedding (nomic-embed-text) runs on CPU — fast enough for search (50-150ms)
-      - OLLAMA_CPU_URL=http://host.docker.internal:11435
-    extra_hosts:
-      - "host.docker.internal:host-gateway"
-    restart: unless-stopped
-
-  grammy:
-    build: ./grammy
-    container_name: grammy
-    ports:
-      - "3001:3001"
-    environment:
-      - TELEGRAM_BOT_TOKEN=${TELEGRAM_BOT_TOKEN}
-      - DEEPAGENTS_URL=http://deepagents:8000
-    restart: unless-stopped
--- a/adolf/openmemory/server.py
+++ b/adolf/openmemory/server.py
@@ -1,62 +0,0 @@
-import os
-from mcp.server.fastmcp import FastMCP
-from mem0 import Memory
-
-OLLAMA_CPU_URL = os.getenv("OLLAMA_CPU_URL", "http://host.docker.internal:11435")
-QDRANT_HOST = os.getenv("QDRANT_HOST", "host.docker.internal")
-QDRANT_PORT = int(os.getenv("QDRANT_PORT", "6333"))
-
-config = {
-    "llm": {
-        "provider": "ollama",
-        "config": {
-            "model": "qwen2.5:1.5b",
-            "ollama_base_url": OLLAMA_CPU_URL,
-        },
-    },
-    "embedder": {
-        "provider": "ollama",
-        "config": {
-            "model": "nomic-embed-text",
-            "ollama_base_url": OLLAMA_CPU_URL,
-        },
-    },
-    "vector_store": {
-        "provider": "qdrant",
-        "config": {
-            "collection_name": "adolf_memories",
-            "embedding_model_dims": 768,
-            "host": QDRANT_HOST,
-            "port": QDRANT_PORT,
-        },
-    },
-}
-
-memory = Memory.from_config(config)
-
-mcp = FastMCP("openmemory", host="0.0.0.0", port=8765)
-
-
-@mcp.tool()
-def add_memory(text: str, user_id: str = "default") -> str:
-    """Store a memory for a user."""
-    result = memory.add(text, user_id=user_id)
-    return str(result)
-
-
-@mcp.tool()
-def search_memory(query: str, user_id: str = "default") -> str:
-    """Search memories for a user using semantic similarity."""
-    results = memory.search(query, user_id=user_id)
-    return str(results)
-
-
-@mcp.tool()
-def get_all_memories(user_id: str = "default") -> str:
-    """Get all stored memories for a user."""
-    results = memory.get_all(user_id=user_id)
-    return str(results)
-
-
-if __name__ == "__main__":
-    mcp.run(transport="sse")
--- a/adolf/potential-directions.md
+++ b/adolf/potential-directions.md
@@ -1,13 +0,0 @@
-# Potential Directions
-
-## CPU Extraction Model Candidates (mem0 / openmemory)
-
-Replacing `gemma3:1b` — documented JSON/structured output failures make it unreliable for mem0's extraction pipeline.
-
-| Rank | Model | Size | CPU speed | JSON reliability | Notes |
-|------|-------|------|-----------|-----------------|-------|
-| 1 | `qwen2.5:1.5b` | ~934 MB | 25–40 tok/s | Excellent | Best fit: fast + structured output, 18T token training |
-| 2 | `qwen2.5:3b` | ~1.9 GB | 15–25 tok/s | Excellent | Quality upgrade, same family |
-| 3 | `llama3.2:3b` | ~2 GB | 15–25 tok/s | Good | Highest IFEval score (77.4) in class |
-| 4 | `smollm2:1.7b` | ~1.1 GB | 25–35 tok/s | Moderate | Use temp=0; NuExtract-1.5-smol is fine-tuned variant |
-| 5 | `phi4-mini` | ~2.5 GB | 10–17 tok/s | Good | Function calling support, borderline CPU speed |
--- a/adolf/router.py
+++ b/adolf/router.py
@@ -1,138 +0,0 @@
-import re
-from typing import Optional
-from langchain_core.messages import SystemMessage, HumanMessage
-
-# ── Regex pre-classifier ──────────────────────────────────────────────────────
-# Catches obvious light-tier patterns before calling the LLM.
-# Keyed by regex → compiled pattern.
-_LIGHT_PATTERNS = re.compile(
-    r"^("
-    # Greetings / farewells
-    r"hi|hello|hey|yo|sup|howdy|good morning|good evening|good night|good afternoon"
-    r"|bye|goodbye|see you|cya|later|ttyl"
-    # Acknowledgements / small talk
-    r"|thanks?|thank you|thx|ty|ok|okay|k|cool|great|awesome|perfect|sounds good|got it|nice|sure"
-    r"|how are you|how are you\?|how are you doing(\s+today)?[?!.]*"
-    r"|what.?s up"
-    # Calendar facts: "what day comes after X?" / "what comes after X?"
-    r"|what\s+day\s+(comes\s+after|follows|is\s+after)\s+\w+[?!.]*"
-    r"|what\s+comes\s+after\s+\w+[?!.]*"
-    # Acronym expansions: "what does X stand for?"
-    r"|what\s+does\s+\w+\s+stand\s+for[?!.]*"
-    r")[\s!.?]*$",
-    re.IGNORECASE,
-)
-
-# ── LLM classification prompt ─────────────────────────────────────────────────
-CLASSIFY_PROMPT = """Classify the message. Output ONLY one word: light, medium, or complex.
-
-LIGHT = answerable from general knowledge, no internet needed:
-  what is 2+2 / what is the capital of France / name the three primary colors
-  tell me a short joke / is the sky blue / is water wet
-
-MEDIUM = requires web search or the user's stored memories:
-  current weather / today's news / Bitcoin price / what did we talk about
-
-COMPLEX = /think prefix only:
-  /think compare frameworks / /think plan a trip
-
-Message: {message}
-Output (one word only — light, medium, or complex):"""
-
-LIGHT_REPLY_PROMPT = """You are a helpful Telegram assistant. Answer briefly and naturally (1-3 sentences). Be friendly."""
-
-
-def _format_history(history: list[dict]) -> str:
-    if not history:
-        return "(none)"
-    lines = []
-    for msg in history:
-        role = msg.get("role", "?")
-        content = str(msg.get("content", ""))[:200]
-        lines.append(f"{role}: {content}")
-    return "\n".join(lines)
-
-
-def _parse_tier(text: str) -> str:
-    """Extract tier from raw model output. Default to medium."""
-    t = text.strip().lower()
-    snippet = t[:60]
-    if "complex" in snippet:
-        return "complex"
-    if "medium" in snippet:
-        return "medium"
-    if "light" in snippet:
-        return "light"
-    # Model invented a descriptive category (e.g. "simplefact", "trivial", "basic") →
-    # treat as light since it recognised the question doesn't need tools
-    if any(w in snippet for w in ("simple", "fact", "trivial", "basic", "easy", "general")):
-        return "light"
-    return "medium"  # safe default
-
-
-class Router:
-    def __init__(self, model):
-        self.model = model
-
-    async def route(
-        self,
-        message: str,
-        history: list[dict],
-        force_complex: bool = False,
-    ) -> tuple[str, Optional[str]]:
-        """
-        Returns (tier, reply_or_None).
-        For light tier: also generates the reply with a second call.
-        For medium/complex: reply is None.
-        """
-        if force_complex:
-            return "complex", None
-
-        # Step 0: regex pre-classification for obvious light patterns
-        if _LIGHT_PATTERNS.match(message.strip()):
-            print(f"[router] regex→light", flush=True)
-            return await self._generate_light_reply(message, history)
-
-        # Step 1: LLM classification with raw text output
-        try:
-            classify_response = await self.model.ainvoke([
-                HumanMessage(content=CLASSIFY_PROMPT.format(message=message)),
-            ])
-            raw = classify_response.content or ""
-            raw = re.sub(r"<think>.*?</think>", "", raw, flags=re.DOTALL).strip()
-            tier = _parse_tier(raw)
-
-            if tier == "complex" and not message.startswith("/think"):
-                tier = "medium"
-
-            print(f"[router] raw={raw[:30]!r} → tier={tier}", flush=True)
-        except Exception as e:
-            print(f"[router] classify error, defaulting to medium: {e}", flush=True)
-            return "medium", None
-
-        if tier != "light":
-            return tier, None
-
-        return await self._generate_light_reply(message, history)
-
-    async def _generate_light_reply(
-        self, message: str, history: list[dict]
-    ) -> tuple[str, Optional[str]]:
-        """Generate a short reply using the router model for light-tier messages."""
-        history_text = _format_history(history)
-        context = f"\nConversation history:\n{history_text}" if history else ""
-        try:
-            reply_response = await self.model.ainvoke([
-                SystemMessage(content=LIGHT_REPLY_PROMPT + context),
-                HumanMessage(content=message),
-            ])
-            reply_text = reply_response.content or ""
-            reply_text = re.sub(r"<think>.*?</think>", "", reply_text, flags=re.DOTALL).strip()
-            if not reply_text:
-                print("[router] light reply empty, falling back to medium", flush=True)
-                return "medium", None
-            print(f"[router] light reply: {len(reply_text)} chars", flush=True)
-            return "light", reply_text
-        except Exception as e:
-            print(f"[router] light reply error, falling back to medium: {e}", flush=True)
-            return "medium", None
--- a/adolf/test_pipeline.py
+++ b/adolf/test_pipeline.py
@@ -1,905 +0,0 @@
-#!/usr/bin/env python3
-"""
-Adolf pipeline integration test with end-to-end timing profiling.
-
-Tests:
-  1. Service health (deepagents, openmemory, grammy MCP SSE)
-  2. GPU Ollama models
-  3. CPU Ollama models
-  4. Qdrant collection + vector dims
-  5. SearXNG
-  6. Name store  — "remember that your name is <RandomName>"
-  7. Qdrant point added after store
-  8. Name recall — "what is your name?" → reply contains <RandomName>
-  9. Timing profile + bottleneck report
- 10. Easy benchmark   — 10 easy questions → all must route to light
- 11. Medium benchmark — 10 medium questions → must route to medium (or light, never complex)
- 12. Hard benchmark   — 10 /think questions → all must route to complex; VRAM flush verified
-
-Usage:
-    python3 test_pipeline.py [--chat-id CHAT_ID]
-                             [--bench-only]       skip sections 1-9, run 10+11+12
-                             [--easy-only]        skip 1-9 and 11+12, run only section 10
-                             [--medium-only]      skip 1-9 and 10+12, run only section 11
-                             [--hard-only]        skip 1-9 and 10+11, run only section 12
-                             [--no-bench]         skip sections 10-12
-
-Timing is extracted from deepagents container logs, not estimated from sleeps.
-"""
-
-import argparse
-import http.client
-import json
-import random
-import re
-import subprocess
-import sys
-import time
-import urllib.request
-
-# ── config ────────────────────────────────────────────────────────────────────
-DEEPAGENTS    = "http://localhost:8000"
-OPENMEMORY    = "http://localhost:8765"
-GRAMMY_HOST   = "localhost"
-GRAMMY_PORT   = 3001
-OLLAMA_GPU    = "http://localhost:11436"
-OLLAMA_CPU    = "http://localhost:11435"
-QDRANT        = "http://localhost:6333"
-SEARXNG       = "http://localhost:11437"
-COMPOSE_FILE  = "/home/alvis/agap_git/adolf/docker-compose.yml"
-DEFAULT_CHAT_ID = "346967270"
-
-NAMES = [
-    "Maximilian", "Cornelius", "Zephyr", "Archibald", "Balthazar",
-    "Ignatius", "Lysander", "Octavian", "Reginald", "Sylvester",
-]
-
-# ── benchmark questions ───────────────────────────────────────────────────────
-BENCHMARK = {
-    "easy": [
-        "hi",
-        "what is 2+2?",
-        "what is the capital of France?",
-        "tell me a short joke",
-        "how are you doing today?",
-        "thanks!",
-        "what day comes after Wednesday?",
-        "name the three primary colors",
-        "is the sky blue?",
-        "what does CPU stand for?",
-    ],
-    "medium": [
-        "what is the current weather in Berlin?",
-        "find the latest news about artificial intelligence",
-        "what is the current price of Bitcoin?",
-        "search for a good pasta carbonara recipe",
-        "what movies are in theaters this week?",
-        "find Python tutorials for beginners",
-        "who won the last FIFA World Cup?",
-        "do you remember what we talked about before?",
-        "search for the best coffee shops in Tokyo",
-        "what is happening in the tech industry this week?",
-    ],
-    "hard": [
-        "/think compare the top 3 Python web frameworks (Django, FastAPI, Flask) and recommend one for a production REST API",
-        "/think research the history of artificial intelligence and create a timeline of key milestones",
-        "/think plan a 7-day trip to Japan with daily itinerary, accommodation suggestions, and budget breakdown",
-        "/think analyze microservices vs monolithic architecture: pros, cons, and when to choose each",
-        "/think write a Python script that reads a CSV file, cleans the data, and generates summary statistics",
-        "/think research quantum computing: explain the key concepts and how it differs from classical computing",
-        "/think compare PostgreSQL, MongoDB, and Redis — when to use each and what are the trade-offs?",
-        "/think create a comprehensive Docker deployment guide covering best practices for production",
-        "/think research climate change: summarize the latest IPCC findings and key data points",
-        "/think design a REST API with authentication, rate limiting, and proper error handling — provide architecture and code outline",
-    ],
-}
-
-PASS = "\033[32mPASS\033[0m"
-FAIL = "\033[31mFAIL\033[0m"
-INFO = "\033[36mINFO\033[0m"
-WARN = "\033[33mWARN\033[0m"
-
-results = []
-timings = {}   # label → float seconds | None
-
-
-# ── helpers ───────────────────────────────────────────────────────────────────
-
-def report(name, ok, detail=""):
-    tag = PASS if ok else FAIL
-    print(f"  [{tag}] {name}" + (f" — {detail}" if detail else ""))
-    results.append((name, ok))
-
-
-def tf(v):
-    """Format timing value."""
-    return f"{v:6.2f}s" if v is not None else "   n/a"
-
-
-def get(url, timeout=5):
-    with urllib.request.urlopen(urllib.request.Request(url), timeout=timeout) as r:
-        return r.status, r.read().decode()
-
-
-def post_json(url, payload, timeout=10):
-    data = json.dumps(payload).encode()
-    req = urllib.request.Request(url, data=data,
-                                  headers={"Content-Type": "application/json"},
-                                  method="POST")
-    with urllib.request.urlopen(req, timeout=timeout) as r:
-        return r.status, json.loads(r.read().decode())
-
-
-def check_sse(host, port, path):
-    try:
-        conn = http.client.HTTPConnection(host, port, timeout=5)
-        conn.request("GET", path, headers={"Accept": "text/event-stream"})
-        r = conn.getresponse()
-        conn.close()
-        return r.status == 200, f"HTTP {r.status}"
-    except Exception as e:
-        return False, str(e)
-
-
-def qdrant_count():
-    try:
-        _, body = get(f"{QDRANT}/collections/adolf_memories")
-        return json.loads(body).get("result", {}).get("points_count", 0)
-    except Exception:
-        return 0
-
-
-def fetch_logs(since_s=600):
-    """Return deepagents log lines from the last since_s seconds."""
-    try:
-        r = subprocess.run(
-            ["docker", "compose", "-f", COMPOSE_FILE, "logs", "deepagents",
-             f"--since={int(since_s)}s", "--no-log-prefix"],
-            capture_output=True, text=True, timeout=15,
-        )
-        return r.stdout.splitlines()
-    except Exception:
-        return []
-
-
-def parse_run_block(lines, msg_prefix):
-    """
-    Scan log lines for the LAST '[agent] running: <msg_prefix>' block.
-    Extracts reply timing, tier, and memory timing from that block.
-
-    Returns dict or None if the reply has not appeared in logs yet.
-    Dict keys:
-      reply_total, llm, send, tier, reply_text  — from "[agent] replied in ..."
-      memory_s                                  — from "[memory] stored in ..."
-      memory_error                              — True if "[memory] error" found
-    """
-    search = msg_prefix[:50]
-    start_idx = None
-    for i, line in enumerate(lines):
-        if "[agent] running:" in line and search in line:
-            start_idx = i  # keep updating — we want the LAST occurrence
-
-    if start_idx is None:
-        return None
-
-    block = lines[start_idx:]
-    last_ai_text = None
-    reply_data = None
-
-    for j, line in enumerate(block):
-        # Track last non-tool AIMessage (the final reply)
-        if "AIMessage:" in line and "→" not in line:
-            txt = line.split("AIMessage:", 1)[-1].strip()
-            if txt:
-                last_ai_text = txt
-
-        # For light tier: router reply is stored in _conversation_buffers directly
-        # so there may be no AIMessage log — grab from tier=light line
-        if "[agent] tier=light" in line and "message=" in line:
-            # Extract preview text logged elsewhere if available
-            pass
-
-        m = re.search(r"replied in ([\d.]+)s \(llm=([\d.]+)s, send=([\d.]+)s\)", line)
-        if m:
-            # Extract optional tier tag at end of line
-            tier_m = re.search(r"\btier=(\w+)", line)
-            tier = tier_m.group(1) if tier_m else "unknown"
-            reply_data = {
-                "reply_total": float(m.group(1)),
-                "llm":         float(m.group(2)),
-                "send":        float(m.group(3)),
-                "tier":        tier,
-                "reply_text":  last_ai_text,
-                "memory_s":    None,
-                "memory_error": False,
-                "_j": j,
-            }
-            break
-
-    if reply_data is None:
-        return None  # reply not in logs yet
-
-    # Memory line can appear after the next "[agent] running:" — no stop condition
-    for line in block[reply_data["_j"] + 1:]:
-        mm = re.search(r"\[memory\] stored in ([\d.]+)s", line)
-        if mm:
-            reply_data["memory_s"] = float(mm.group(1))
-            break
-        if "[memory] error" in line:
-            reply_data["memory_error"] = True
-            break
-
-    return reply_data
-
-
-def wait_for(label, msg_prefix, timeout_s=200, need_memory=True):
-    """
-    Poll deepagents logs until the message is fully processed.
-    Shows a live progress line.
-    Returns timing dict or None on timeout.
-    """
-    t_start = time.monotonic()
-    deadline = t_start + timeout_s
-    tick = 0
-    last_result = None
-
-    while time.monotonic() < deadline:
-        # Window grows with elapsed time — never miss a line that appeared late
-        since = int(time.monotonic() - t_start) + 90
-        lines = fetch_logs(since_s=since)
-        result = parse_run_block(lines, msg_prefix)
-
-        if result:
-            last_result = result
-            has_mem = result["memory_s"] is not None or result["memory_error"]
-            if (not need_memory) or has_mem:
-                elapsed = time.monotonic() - t_start
-                print(f"\r  [{label}] done after {elapsed:.0f}s{' ' * 30}")
-                return result
-
-        time.sleep(4)
-        tick += 1
-        rem = int(deadline - time.monotonic())
-        if last_result:
-            phase = "waiting for memory..." if need_memory else "done"
-        else:
-            phase = "waiting for LLM reply..."
-        print(f"\r  [{label}] {tick*4}s elapsed, {rem}s left — {phase}  ", end="", flush=True)
-
-    print(f"\r  [{label}] TIMEOUT after {timeout_s}s{' ' * 30}")
-    return None
-
-
-# ── args ──────────────────────────────────────────────────────────────────────
-parser = argparse.ArgumentParser(description="Adolf pipeline test")
-parser.add_argument("--chat-id", default=DEFAULT_CHAT_ID)
-parser.add_argument("--bench-only", action="store_true",
-                    help="Skip sections 1-9, run sections 10+11 (both benchmarks)")
-parser.add_argument("--easy-only", action="store_true",
-                    help="Skip sections 1-9 and 11, run only section 10 (easy benchmark)")
-parser.add_argument("--medium-only", action="store_true",
-                    help="Skip sections 1-9 and 10, run only section 11 (medium benchmark)")
-parser.add_argument("--hard-only", action="store_true",
-                    help="Skip sections 1-9 and 10+11, run only section 12 (hard benchmark)")
-parser.add_argument("--no-bench", action="store_true",
-                    help="Skip sections 10-12 (all benchmarks)")
-args = parser.parse_args()
-CHAT_ID = args.chat_id
-
-# Derived flags for readability
-_skip_pipeline = args.bench_only or args.easy_only or args.medium_only or args.hard_only
-_run_easy   = not args.no_bench and not args.medium_only and not args.hard_only
-_run_medium = not args.no_bench and not args.easy_only   and not args.hard_only
-_run_hard   = not args.no_bench and not args.easy_only   and not args.medium_only
-
-random_name = random.choice(NAMES)
-
-if not _skip_pipeline:
-    print(f"\n  Test name : \033[1m{random_name}\033[0m")
-    print(f"  Chat ID   : {CHAT_ID}")
-
-
-# ── 1. service health ─────────────────────────────────────────────────────────
-if not _skip_pipeline:
-    print(f"\n[{INFO}] 1. Service health")
-    t0 = time.monotonic()
-
-    try:
-        status, body = get(f"{DEEPAGENTS}/health")
-        data = json.loads(body)
-        ok = status == 200 and data.get("agent_ready") is True
-        report("deepagents /health — agent_ready", ok, f"agent_ready={data.get('agent_ready')}")
-    except Exception as e:
-        report("deepagents /health", False, str(e))
-
-    ok, detail = check_sse("localhost", 8765, "/sse")
-    report("openmemory /sse reachable", ok, detail)
-
-    ok, detail = check_sse(GRAMMY_HOST, GRAMMY_PORT, "/sse")
-    report("grammy /sse reachable", ok, detail)
-
-    timings["health_check"] = time.monotonic() - t0
-
-
-# ── 2. GPU Ollama ─────────────────────────────────────────────────────────────
-if not _skip_pipeline:
-    print(f"\n[{INFO}] 2. GPU Ollama (port 11436)")
-    t0 = time.monotonic()
-
-    try:
-        status, body = get(f"{OLLAMA_GPU}/api/tags")
-        models = [m["name"] for m in json.loads(body).get("models", [])]
-        has_qwen = any("qwen3" in m for m in models)
-        report("GPU Ollama reachable", True, f"models: {models}")
-        report("qwen3:8b present", has_qwen)
-    except Exception as e:
-        report("GPU Ollama reachable", False, str(e))
-        report("qwen3:8b present", False, "skipped")
-
-    timings["gpu_ollama_ping"] = time.monotonic() - t0
-
-
-# ── 3. CPU Ollama ─────────────────────────────────────────────────────────────
-if not _skip_pipeline:
-    print(f"\n[{INFO}] 3. CPU Ollama (port 11435)")
-    t0 = time.monotonic()
-
-    try:
-        status, body = get(f"{OLLAMA_CPU}/api/tags")
-        models = [m["name"] for m in json.loads(body).get("models", [])]
-        has_embed = any("nomic-embed-text" in m for m in models)
-        report("CPU Ollama reachable", True, f"models: {models}")
-        report("nomic-embed-text present", has_embed)
-    except Exception as e:
-        report("CPU Ollama reachable", False, str(e))
-        report("nomic-embed-text present", False, "skipped")
-
-    timings["cpu_ollama_ping"] = time.monotonic() - t0
-
-
-# ── 4. Qdrant ─────────────────────────────────────────────────────────────────
-if not _skip_pipeline:
-    print(f"\n[{INFO}] 4. Qdrant (port 6333)")
-    t0 = time.monotonic()
-
-    try:
-        status, body = get(f"{QDRANT}/collections")
-        cols = [c["name"] for c in json.loads(body).get("result", {}).get("collections", [])]
-        report("Qdrant reachable", True, f"collections: {cols}")
-        report("adolf_memories collection exists", "adolf_memories" in cols)
-    except Exception as e:
-        report("Qdrant reachable", False, str(e))
-        report("adolf_memories collection exists", False, "skipped")
-
-    try:
-        status, body = get(f"{QDRANT}/collections/adolf_memories")
-        info = json.loads(body).get("result", {})
-        dims = info.get("config", {}).get("params", {}).get("vectors", {}).get("size")
-        report("vector dims = 768", dims == 768, f"got {dims}")
-    except Exception as e:
-        report("adolf_memories collection info", False, str(e))
-
-    timings["qdrant_ping"] = time.monotonic() - t0
-
-
-# ── 5. SearXNG ────────────────────────────────────────────────────────────────
-if not _skip_pipeline:
-    print(f"\n[{INFO}] 5. SearXNG (port 11437)")
-    t0 = time.monotonic()
-
-    try:
-        status, body = get(f"{SEARXNG}/search?q=test&format=json", timeout=15)
-        elapsed = time.monotonic() - t0
-        n = len(json.loads(body).get("results", []))
-        report("SearXNG reachable + JSON results", status == 200 and n > 0, f"{n} results in {elapsed:.1f}s")
-        report("SearXNG response < 5s", elapsed < 5, f"{elapsed:.2f}s")
-        timings["searxng_latency"] = elapsed
-    except Exception as e:
-        report("SearXNG reachable", False, str(e))
-        report("SearXNG response < 5s", False, "skipped")
-        timings["searxng_latency"] = None
-
-    timings["searxng_check"] = time.monotonic() - t0
-
-
-# ── 6–8. Name memory pipeline ─────────────────────────────────────────────────
-if not _skip_pipeline:
-    print(f"\n[{INFO}] 6–8. Name memory pipeline")
-    print(f"  chat_id={CHAT_ID}  name={random_name}")
-
-    store_msg  = f"remember that your name is {random_name}"
-    recall_msg = "what is your name?"
-
-    pts_before = qdrant_count()
-    print(f"  Qdrant points before: {pts_before}")
-
-    # ── 6. Send store message ─────────────────────────────────────────────────────
-    print(f"\n  [store] '{store_msg}'")
-    t_store = time.monotonic()
-
-    try:
-        status, _ = post_json(f"{DEEPAGENTS}/chat",
-                               {"message": store_msg, "chat_id": CHAT_ID}, timeout=5)
-        t_accept = time.monotonic() - t_store
-        report("POST /chat (store) returns 202 immediately",
-               status == 202 and t_accept < 1, f"status={status}, t={t_accept:.3f}s")
-        timings["store_http_accept"] = t_accept
-    except Exception as e:
-        report("POST /chat (store)", False, str(e))
-        sys.exit(1)
-
-    store = wait_for("store", store_msg, timeout_s=220, need_memory=True)
-
-    if store:
-        timings["store_llm"]    = store["llm"]
-        timings["store_send"]   = store["send"]
-        timings["store_reply"]  = store["reply_total"]
-        timings["store_memory"] = store["memory_s"]
-        report("Agent replied to store message", True,
-               f"{store['reply_total']:.1f}s total  llm={store['llm']:.1f}s  send={store['send']:.1f}s  tier={store['tier']}")
-        if store["memory_s"] is not None:
-            report("Memory stored without error", True, f"{store['memory_s']:.1f}s")
-        elif store["memory_error"]:
-            report("Memory stored without error", False, "error in [memory] log")
-        else:
-            report("Memory stored without error", False, "not found in logs (still running?)")
-        print(f"    Store reply: {store['reply_text']!r}")
-    else:
-        report("Agent replied to store message", False, "timeout")
-        report("Memory stored without error", False, "timeout")
-        sys.exit(1)
-
-    # ── 7. Verify Qdrant ──────────────────────────────────────────────────────────
-    pts_after = qdrant_count()
-    new_pts = pts_after - pts_before
-    report("New memory point(s) added to Qdrant", new_pts > 0,
-           f"{pts_before} → {pts_after} (+{new_pts})")
-    timings["qdrant_new_points"] = new_pts
-
-    # ── 8. Send recall message ────────────────────────────────────────────────────
-    print(f"\n  [recall] '{recall_msg}'")
-    t_recall = time.monotonic()
-
-    try:
-        status, _ = post_json(f"{DEEPAGENTS}/chat",
-                               {"message": recall_msg, "chat_id": CHAT_ID}, timeout=5)
-        t_accept2 = time.monotonic() - t_recall
-        report("POST /chat (recall) returns 202 immediately",
-               status == 202 and t_accept2 < 1, f"status={status}, t={t_accept2:.3f}s")
-        timings["recall_http_accept"] = t_accept2
-    except Exception as e:
-        report("POST /chat (recall)", False, str(e))
-
-    recall = wait_for("recall", recall_msg, timeout_s=160, need_memory=False)
-
-    if recall:
-        timings["recall_llm"]   = recall["llm"]
-        timings["recall_send"]  = recall["send"]
-        timings["recall_reply"] = recall["reply_total"]
-        report("Agent replied to recall message", True,
-               f"{recall['reply_total']:.1f}s total  llm={recall['llm']:.1f}s  send={recall['send']:.1f}s  tier={recall['tier']}")
-        reply_text = recall["reply_text"] or ""
-        name_in_reply = random_name.lower() in reply_text.lower()
-        report(f"Reply contains '{random_name}'", name_in_reply,
-               f"reply: {reply_text[:120]!r}")
-    else:
-        report("Agent replied to recall message", False, "timeout")
-        report(f"Reply contains '{random_name}'", False, "no reply")
-
-
-# ── 9. Timing profile ─────────────────────────────────────────────────────────
-if not _skip_pipeline:
-    print(f"\n[{INFO}] 9. Timing profile")
-
-    W = 36
-
-    print(f"\n  {'Stage':<{W}}  {'Time':>8}")
-    print(f"  {'─'*W}  {'─'*8}")
-
-    rows_store = [
-        ("[GPU] HTTP accept — store turn",       "store_http_accept"),
-        ("[GPU] qwen3:Xb inference — store turn","store_llm"),
-        ("[GPU] Telegram send — store turn",     "store_send"),
-        ("[GPU] Total reply latency — store",    "store_reply"),
-        ("[GPU] qwen2.5:1.5b+embed — async mem", "store_memory"),
-    ]
-    rows_recall = [
-        ("[GPU] HTTP accept — recall turn",      "recall_http_accept"),
-        ("[GPU] qwen3:Xb inference — recall",    "recall_llm"),
-        ("[GPU] Telegram send — recall turn",    "recall_send"),
-        ("[GPU] Total reply latency — recall",   "recall_reply"),
-    ]
-
-    for label, key in rows_store:
-        v = timings.get(key)
-        print(f"  {label:<{W}}  {tf(v):>8}")
-
-    print(f"  {'─'*W}  {'─'*8}")
-
-    for label, key in rows_recall:
-        v = timings.get(key)
-        print(f"  {label:<{W}}  {tf(v):>8}")
-
-    # Bottleneck bar chart
-    print(f"\n  Bottleneck analysis (each █ ≈ 5s):")
-    print(f"  {'─'*(W+12)}")
-
-    candidates = [
-        ("[GPU] qwen3:Xb — store reply ", timings.get("store_llm")      or 0),
-        ("[GPU] qwen3:Xb — recall reply", timings.get("recall_llm")     or 0),
-        ("[GPU] qwen2.5:1.5b+embed (async)", timings.get("store_memory")  or 0),
-        ("[net] SearXNG               ",  timings.get("searxng_latency") or 0),
-    ]
-    candidates.sort(key=lambda x: x[1], reverse=True)
-
-    for label, t in candidates:
-        bar = "█" * min(int(t / 5), 24)
-        pct = ""
-        total_pipeline = (timings.get("store_reply") or 0) + (timings.get("store_memory") or 0)
-        if total_pipeline > 0:
-            pct = f"  {t/total_pipeline*100:4.0f}%"
-        print(f"  {label}  {t:6.1f}s  {bar}{pct}")
-
-    print()
-
-
-# ── 10. Tier routing benchmark — easy questions → light path ──────────────────
-if _run_easy:
-    print(f"\n[{INFO}] 10. Tier routing benchmark")
-    print(f"  Sending {len(BENCHMARK['easy'])} easy questions — all must route to 'light'")
-    print(f"  Chat ID: {CHAT_ID}")
-    print()
-
-    bench_results = []   # list of (question, tier, latency_s, ok)
-    LIGHT_TIMEOUT = 60   # seconds — light is fast but may queue behind prior messages
-
-    for i, question in enumerate(BENCHMARK["easy"], 1):
-        tag = f"easy-{i:02d}"
-        short_q = question[:55]
-        print(f"  [{tag}] {short_q!r}")
-
-        # Send
-        t_send = time.monotonic()
-        try:
-            status, _ = post_json(f"{DEEPAGENTS}/chat",
-                                   {"message": question, "chat_id": CHAT_ID}, timeout=5)
-            if status != 202:
-                print(f"          → [{FAIL}] POST returned {status}")
-                bench_results.append((question, "?", None, False))
-                continue
-        except Exception as e:
-            print(f"          → [{FAIL}] POST error: {e}")
-            bench_results.append((question, "?", None, False))
-            continue
-
-        # Poll for reply
-        t_start = time.monotonic()
-        found = None
-        while time.monotonic() - t_start < LIGHT_TIMEOUT:
-            since = int(time.monotonic() - t_start) + 30
-            lines = fetch_logs(since_s=since)
-            found = parse_run_block(lines, question)
-            if found:
-                break
-            time.sleep(1)
-
-        elapsed = time.monotonic() - t_send
-
-        if not found:
-            print(f"          → [{FAIL}] no reply within {LIGHT_TIMEOUT}s")
-            bench_results.append((question, "timeout", None, False))
-            continue
-
-        tier = found.get("tier", "unknown")
-        is_light = (tier == "light")
-        tag_str = PASS if is_light else FAIL
-        print(f"          → [{tag_str}] tier={tier}  latency={found['reply_total']:.1f}s  llm={found['llm']:.1f}s")
-        bench_results.append((question, tier, found["reply_total"], is_light))
-
-        # Brief pause between questions to keep logs clean
-        time.sleep(1)
-
-    # Summary table
-    print(f"\n  {'#':<4}  {'Tier':<8}  {'Latency':>8}  {'Question'}")
-    print(f"  {'─'*4}  {'─'*8}  {'─'*8}  {'─'*50}")
-    for idx, (q, tier, lat, ok) in enumerate(bench_results, 1):
-        lat_str = f"{lat:.1f}s" if lat is not None else "timeout"
-        ok_str = "✓" if ok else "✗"
-        print(f"  {ok_str} {idx:<3}  {tier:<8}  {lat_str:>8}  {q[:50]!r}")
-
-    light_count = sum(1 for _, _, _, ok in bench_results if ok)
-    total_bench = len(bench_results)
-    lats = [lat for _, _, lat, ok in bench_results if ok and lat is not None]
-    avg_lat = sum(lats) / len(lats) if lats else 0
-
-    print(f"\n  Light-path score: {light_count}/{total_bench}")
-    if lats:
-        print(f"  Avg latency (light): {avg_lat:.1f}s  "
-              f"min={min(lats):.1f}s  max={max(lats):.1f}s")
-
-    report(f"All easy questions routed to light ({light_count}/{total_bench})",
-           light_count == total_bench,
-           f"{light_count}/{total_bench} via light path, avg {avg_lat:.1f}s")
-
-
-# ── 11. Medium benchmark — medium questions → medium or light, never complex ──
-if _run_medium:
-    print(f"\n[{INFO}] 11. Medium routing benchmark")
-    print(f"  Sending {len(BENCHMARK['medium'])} medium questions")
-    print(f"  Expected: tier=medium (needs tools). Light is acceptable for factual questions.")
-    print(f"  Fail condition: tier=complex or timeout.")
-    print(f"  Chat ID: {CHAT_ID}")
-    print()
-
-    # Questions where light is a valid alternative (model may know from training data)
-    LIGHT_ACCEPTABLE = {
-        "who won the last FIFA World Cup?",
-        "search for a good pasta carbonara recipe",
-        "find Python tutorials for beginners",
-        "search for the best coffee shops in Tokyo",
-    }
-
-    med_results = []   # list of (question, tier, latency_s, correct)
-    MEDIUM_TIMEOUT = 120  # seconds — medium takes 20-100s, allow for queue buildup
-
-    for i, question in enumerate(BENCHMARK["medium"], 1):
-        tag = f"med-{i:02d}"
-        short_q = question[:60]
-        print(f"  [{tag}] {short_q!r}")
-
-        # Send
-        t_send = time.monotonic()
-        try:
-            status, _ = post_json(f"{DEEPAGENTS}/chat",
-                                   {"message": question, "chat_id": CHAT_ID}, timeout=5)
-            if status != 202:
-                print(f"          → [{FAIL}] POST returned {status}")
-                med_results.append((question, "?", None, False))
-                continue
-        except Exception as e:
-            print(f"          → [{FAIL}] POST error: {e}")
-            med_results.append((question, "?", None, False))
-            continue
-
-        # Poll for reply
-        t_start = time.monotonic()
-        found = None
-        while time.monotonic() - t_start < MEDIUM_TIMEOUT:
-            since = int(time.monotonic() - t_start) + 60
-            lines = fetch_logs(since_s=since)
-            found = parse_run_block(lines, question)
-            if found:
-                break
-            time.sleep(3)
-
-        elapsed = time.monotonic() - t_send
-
-        if not found:
-            print(f"          → [{FAIL}] no reply within {MEDIUM_TIMEOUT}s")
-            med_results.append((question, "timeout", None, False))
-            continue
-
-        tier = found.get("tier", "unknown")
-        light_ok = question in LIGHT_ACCEPTABLE
-
-        if tier == "medium":
-            correct = True
-            label = PASS
-            note = "medium ✓"
-        elif tier == "light":
-            correct = light_ok  # light is only acceptable for certain questions
-            label = WARN if not light_ok else PASS
-            note = "light (acceptable)" if light_ok else "light (should be medium)"
-        elif tier == "complex":
-            correct = False
-            label = FAIL
-            note = "complex — wrong escalation"
-        else:
-            correct = False
-            label = FAIL
-            note = f"unknown tier {tier!r}"
-
-        print(f"          → [{label}] {note}  latency={found['reply_total']:.1f}s  llm={found['llm']:.1f}s")
-        med_results.append((question, tier, found["reply_total"], correct))
-
-        # Brief pause between questions
-        time.sleep(1)
-
-    # Summary table
-    print(f"\n  {'#':<4}  {'Tier':<8}  {'Latency':>8}  {'Question'}")
-    print(f"  {'─'*4}  {'─'*8}  {'─'*8}  {'─'*55}")
-    for idx, (q, tier, lat, ok) in enumerate(med_results, 1):
-        lat_str = f"{lat:.1f}s" if lat is not None else "timeout"
-        ok_str = "✓" if ok else ("~" if tier == "light" else "✗")
-        print(f"  {ok_str} {idx:<3}  {tier:<8}  {lat_str:>8}  {q[:55]!r}")
-
-    total_med     = len(med_results)
-    medium_count  = sum(1 for _, tier, _, _ in med_results if tier == "medium")
-    light_count   = sum(1 for _, tier, _, _ in med_results if tier == "light")
-    complex_count = sum(1 for _, tier, _, _ in med_results if tier == "complex")
-    timeout_count = sum(1 for _, tier, _, _ in med_results if tier == "timeout")
-    light_misroute = sum(
-        1 for q, tier, _, _ in med_results
-        if tier == "light" and q not in LIGHT_ACCEPTABLE
-    )
-    lats = [lat for _, _, lat, _ in med_results if lat is not None]
-    correct_count = medium_count + (light_count - light_misroute)
-
-    print(f"\n  Breakdown: medium={medium_count}  light={light_count}  complex={complex_count}  timeout={timeout_count}")
-    if light_misroute:
-        print(f"  [{WARN}] {light_misroute} question(s) answered via light when medium expected (check reply quality)")
-    if lats:
-        print(f"  Avg latency: {sum(lats)/len(lats):.1f}s  min={min(lats):.1f}s  max={max(lats):.1f}s")
-
-    no_complex   = complex_count == 0
-    no_timeout   = timeout_count == 0
-    all_ok = no_complex and no_timeout
-
-    report(
-        f"Medium questions: no complex escalation ({medium_count + light_count}/{total_med} routed)",
-        no_complex,
-        f"medium={medium_count} light={light_count} complex={complex_count} timeout={timeout_count}",
-    )
-    if not no_timeout:
-        report(
-            f"Medium questions: all completed within {MEDIUM_TIMEOUT}s",
-            False,
-            f"{timeout_count} question(s) timed out (increase MEDIUM_TIMEOUT or check agent logs)",
-        )
-
-
-# ── 12. Hard benchmark — /think questions → complex tier + VRAM flush verified ─
-if _run_hard:
-    print(f"\n[{INFO}] 12. Hard routing benchmark")
-    print(f"  Sending {len(BENCHMARK['hard'])} /think questions — all must route to 'complex'")
-    print(f"  Verifies: /think prefix → force_complex=True → VRAM flush → qwen3:8b inference")
-    print(f"  Acceptable fallback: 'medium' if VRAM eviction timed out (logged warning)")
-    print(f"  Fail condition: tier=light or timeout")
-    print(f"  Chat ID: {CHAT_ID}")
-    print()
-
-    hard_results  = []    # list of (question, tier, latency_s, ok)
-    COMPLEX_TIMEOUT = 300  # seconds — complex takes 60-180s + VRAM flush overhead
-
-    # Log markers we expect to see for complex path
-    _VRAM_ENTER = "[vram] enter_complex_mode"
-    _VRAM_EXIT  = "[vram] exit_complex_mode"
-
-    for i, question in enumerate(BENCHMARK["hard"], 1):
-        tag = f"hard-{i:02d}"
-        # Strip /think prefix for display
-        short_q = question[len("/think "):].strip()[:60]
-        print(f"  [{tag}] /think {short_q!r}")
-
-        # Snapshot log window start time
-        t_send = time.monotonic()
-        try:
-            status, _ = post_json(f"{DEEPAGENTS}/chat",
-                                   {"message": question, "chat_id": CHAT_ID}, timeout=5)
-            if status != 202:
-                print(f"          → [{FAIL}] POST returned {status}")
-                hard_results.append((question, "?", None, False))
-                continue
-        except Exception as e:
-            print(f"          → [{FAIL}] POST error: {e}")
-            hard_results.append((question, "?", None, False))
-            continue
-
-        # Poll for reply
-        t_start = time.monotonic()
-        found = None
-        while time.monotonic() - t_start < COMPLEX_TIMEOUT:
-            since = int(time.monotonic() - t_start) + 90
-            lines = fetch_logs(since_s=since)
-            found = parse_run_block(lines, question[len("/think "):].strip())
-            if found:
-                break
-            time.sleep(5)
-
-        elapsed = time.monotonic() - t_send
-
-        if not found:
-            print(f"          → [{FAIL}] no reply within {COMPLEX_TIMEOUT}s")
-            hard_results.append((question, "timeout", None, False))
-            continue
-
-        tier = found.get("tier", "unknown")
-
-        if tier == "complex":
-            ok = True
-            label = PASS
-            note = "complex ✓"
-        elif tier == "medium":
-            # Acceptable fallback if VRAM eviction timed out
-            ok = True
-            label = WARN
-            note = "medium (VRAM fallback — check [vram] logs)"
-        else:
-            ok = False
-            label = FAIL
-            note = f"tier={tier} — unexpected"
-
-        # Check if VRAM enter/exit were logged for this block
-        lines_block = fetch_logs(since_s=int(elapsed) + 120)
-        msg_key = question[len("/think "):].strip()[:40]
-        vram_enter_seen = any(_VRAM_ENTER in ln for ln in lines_block
-                              if msg_key in ln or
-                              any(msg_key[:15] in prev_ln
-                                  for prev_ln in lines_block[max(0, lines_block.index(ln)-10):lines_block.index(ln)]))
-
-        # Simpler: just check the recent log window for enter/exit markers
-        recent = "\n".join(lines_block[-200:])
-        vram_enter_seen = _VRAM_ENTER in recent
-        vram_exit_seen  = _VRAM_EXIT  in recent
-
-        vram_note = ""
-        if tier == "complex":
-            if vram_enter_seen:
-                vram_note = " [vram:flush✓]"
-            else:
-                vram_note = f" [{WARN}:no vram flush log]"
-
-        print(f"          → [{label}] {note}  latency={found['reply_total']:.1f}s  llm={found['llm']:.1f}s{vram_note}")
-        hard_results.append((question, tier, found["reply_total"], ok))
-
-        # Pause to let exit_complex_mode background task complete before next question
-        # (flushes qwen3:8b and pre-warms 4b+router — avoids VRAM conflict on next enter)
-        time.sleep(5)
-
-    # Summary table
-    print(f"\n  {'#':<4}  {'Tier':<8}  {'Latency':>8}  {'Question (/think ...)'}")
-    print(f"  {'─'*4}  {'─'*8}  {'─'*8}  {'─'*55}")
-    for idx, (q, tier, lat, ok) in enumerate(hard_results, 1):
-        lat_str = f"{lat:.1f}s" if lat is not None else "timeout"
-        ok_str = "✓" if tier == "complex" else ("~" if tier == "medium" else "✗")
-        short = q[len("/think "):].strip()[:55]
-        print(f"  {ok_str} {idx:<3}  {tier:<8}  {lat_str:>8}  {short!r}")
-
-    total_hard   = len(hard_results)
-    complex_count = sum(1 for _, t, _, _ in hard_results if t == "complex")
-    medium_fb    = sum(1 for _, t, _, _ in hard_results if t == "medium")
-    light_count  = sum(1 for _, t, _, _ in hard_results if t == "light")
-    timeout_count = sum(1 for _, t, _, _ in hard_results if t == "timeout")
-    lats = [lat for _, _, lat, _ in hard_results if lat is not None]
-
-    print(f"\n  Breakdown: complex={complex_count}  medium(fallback)={medium_fb}  light={light_count}  timeout={timeout_count}")
-    if medium_fb:
-        print(f"  [{WARN}] {medium_fb} question(s) fell back to medium (VRAM eviction timeout)")
-    if light_count:
-        print(f"  [{FAIL}] {light_count} question(s) routed to light — /think prefix not detected")
-    if lats:
-        print(f"  Avg latency: {sum(lats)/len(lats):.1f}s  min={min(lats):.1f}s  max={max(lats):.1f}s")
-
-    no_light    = light_count == 0
-    no_timeout  = timeout_count == 0
-
-    report(
-        f"Hard questions routed to complex (not light) ({complex_count + medium_fb}/{total_hard})",
-        no_light and no_timeout,
-        f"complex={complex_count} medium_fallback={medium_fb} light={light_count} timeout={timeout_count}",
-    )
-
-
-# ── summary ───────────────────────────────────────────────────────────────────
-print(f"\n{'─'*55}")
-total  = len(results)
-passed = sum(1 for _, ok in results if ok)
-failed = total - passed
-print(f"Results: {passed}/{total} passed", end="")
-if failed:
-    print(f"  ({failed} failed)\n")
-    print("Failed checks:")
-    for name, ok in results:
-        if not ok:
-            print(f"  - {name}")
-else:
-    print(" — all good")
-print()
-
-# Print benchmark reference
-print(f"[{INFO}] Benchmark questions reference:")
-for tier_name, questions in BENCHMARK.items():
-    print(f"\n  {tier_name.upper()} ({len(questions)} questions):")
-    for j, q in enumerate(questions, 1):
-        print(f"    {j:2d}. {q}")
-print()
--- a/adolf/vram_manager.py
+++ b/adolf/vram_manager.py
@@ -1,71 +0,0 @@
-import asyncio
-import os
-import httpx
-
-OLLAMA_BASE_URL = os.getenv("OLLAMA_BASE_URL", "http://localhost:11434")
-
-
-class VRAMManager:
-    MEDIUM_MODELS = ["qwen3:4b", "qwen2.5:1.5b"]
-    COMPLEX_MODEL = "qwen3:8b"
-
-    def __init__(self, base_url: str = OLLAMA_BASE_URL):
-        self.base_url = base_url
-
-    async def enter_complex_mode(self) -> bool:
-        """Flush medium models before loading 8b. Returns False if eviction timed out."""
-        print("[vram] enter_complex_mode: flushing medium models", flush=True)
-        await asyncio.gather(*[self._flush(m) for m in self.MEDIUM_MODELS])
-        ok = await self._poll_evicted(self.MEDIUM_MODELS, timeout=15)
-        if ok:
-            print("[vram] enter_complex_mode: eviction confirmed, loading qwen3:8b", flush=True)
-        else:
-            print("[vram] enter_complex_mode: eviction timeout — falling back to medium", flush=True)
-        return ok
-
-    async def exit_complex_mode(self):
-        """Flush 8b and pre-warm medium models. Run as background task after complex reply."""
-        print("[vram] exit_complex_mode: flushing qwen3:8b", flush=True)
-        await self._flush(self.COMPLEX_MODEL)
-        print("[vram] exit_complex_mode: pre-warming medium models", flush=True)
-        await asyncio.gather(*[self._prewarm(m) for m in self.MEDIUM_MODELS])
-        print("[vram] exit_complex_mode: done", flush=True)
-
-    async def _flush(self, model: str):
-        """Send keep_alive=0 to force immediate unload from VRAM."""
-        try:
-            async with httpx.AsyncClient(timeout=10.0) as client:
-                await client.post(
-                    f"{self.base_url}/api/generate",
-                    json={"model": model, "prompt": "", "keep_alive": 0},
-                )
-        except Exception as e:
-            print(f"[vram] flush {model} error: {e}", flush=True)
-
-    async def _poll_evicted(self, models: list[str], timeout: float) -> bool:
-        """Poll /api/ps until none of the given models appear (or timeout)."""
-        deadline = asyncio.get_event_loop().time() + timeout
-        while asyncio.get_event_loop().time() < deadline:
-            try:
-                async with httpx.AsyncClient(timeout=5.0) as client:
-                    resp = await client.get(f"{self.base_url}/api/ps")
-                    data = resp.json()
-                    loaded = {m.get("name", "") for m in data.get("models", [])}
-                    if not any(m in loaded for m in models):
-                        return True
-            except Exception as e:
-                print(f"[vram] poll_evicted error: {e}", flush=True)
-            await asyncio.sleep(0.5)
-        return False
-
-    async def _prewarm(self, model: str):
-        """Load model into VRAM with keep_alive=300 (5 min)."""
-        try:
-            async with httpx.AsyncClient(timeout=60.0) as client:
-                await client.post(
-                    f"{self.base_url}/api/generate",
-                    json={"model": model, "prompt": "", "keep_alive": 300},
-                )
-                print(f"[vram] pre-warmed {model}", flush=True)
-        except Exception as e:
-            print(f"[vram] prewarm {model} error: {e}", flush=True)
--- a/haos/CLAUDE.md
+++ b/haos/CLAUDE.md
@@ -0,0 +1,191 @@
+# Home Assistant REST API
+
+## Connection
+
+- **Base URL**: `http://<HA_IP>:8123/api/`
+- **Auth header**: `Authorization: Bearer <TOKEN>`
+- **Token**: Generate at `http://<HA_IP>:8123/profile` → Long-Lived Access Tokens
+- **Response format**: JSON (except `/api/error_log` which is plaintext)
+
+Store token in env var, never hardcode:
+```bash
+export HA_TOKEN="your_token_here"
+export HA_URL="http://<HA_IP>:8123"
+```
+
+## Status Codes
+
+| Code | Meaning |
+|------|---------|
+| 200 | Success (existing resource) |
+| 201 | Created (new resource) |
+| 400 | Bad request |
+| 401 | Unauthorized |
+| 404 | Not found |
+| 405 | Method not allowed |
+
+## GET Endpoints
+
+```bash
+# Health check
+GET /api/
+
+# Current HA configuration
+GET /api/config
+
+# Loaded components
+GET /api/components
+
+# All entity states
+GET /api/states
+
+# Specific entity state
+GET /api/states/<entity_id>
+
+# Available services
+GET /api/services
+
+# Available events
+GET /api/events
+
+# Error log (plaintext)
+GET /api/error_log
+
+# Camera image
+GET /api/camera_proxy/<camera_entity_id>
+
+# All calendar entities
+GET /api/calendars
+
+# Calendar events (start and end are required ISO timestamps)
+GET /api/calendars/<calendar_entity_id>?start=<ISO>&end=<ISO>
+
+# Historical state changes
+GET /api/history/period/<ISO_timestamp>?filter_entity_id=<entity_id>
+# Optional params: end_time, minimal_response, no_attributes, significant_changes_only
+
+# Logbook entries
+GET /api/logbook/<ISO_timestamp>
+# Optional params: entity=<entity_id>, end_time=<ISO>
+```
+
+## POST Endpoints
+
+```bash
+# Create or update entity state (virtual, not device)
+POST /api/states/<entity_id>
+{"state": "on", "attributes": {"brightness": 255}}
+
+# Fire an event
+POST /api/events/<event_type>
+{"optional": "event_data"}
+
+# Call a service
+POST /api/services/<domain>/<service>
+{"entity_id": "light.living_room"}
+
+# Call service and get its response
+POST /api/services/<domain>/<service>?return_response
+{"entity_id": "..."}
+
+# Render a Jinja2 template
+POST /api/template
+{"template": "{{ states('sensor.temperature') }}"}
+
+# Validate configuration
+POST /api/config/core/check_config
+
+# Handle an intent
+POST /api/intent/handle
+{"name": "HassTurnOn", "data": {"name": "lights"}}
+```
+
+## DELETE Endpoints
+
+```bash
+# Remove an entity
+DELETE /api/states/<entity_id>
+```
+
+## Example curl Usage
+
+```bash
+# Health check
+curl -s -H "Authorization: Bearer $HA_TOKEN" $HA_URL/api/
+
+# Get all states
+curl -s -H "Authorization: Bearer $HA_TOKEN" $HA_URL/api/states | jq .
+
+# Get specific entity
+curl -s -H "Authorization: Bearer $HA_TOKEN" $HA_URL/api/states/light.living_room
+
+# Turn on a light
+curl -s -X POST \
+  -H "Authorization: Bearer $HA_TOKEN" \
+  -H "Content-Type: application/json" \
+  -d '{"entity_id": "light.living_room"}' \
+  $HA_URL/api/services/light/turn_on
+
+# Render template
+curl -s -X POST \
+  -H "Authorization: Bearer $HA_TOKEN" \
+  -H "Content-Type: application/json" \
+  -d '{"template": "{{ states(\"sensor.temperature\") }}"}' \
+  $HA_URL/api/template
+```
+
+## Devices
+
+### Lights
+4x Zigbee Tuya lights (TZ3210 TS0505B):
+- `light.tz3210_r5afgmkl_ts0505b` (G2)
+- `light.tz3210_r5afgmkl_ts0505b_g2` (G22)
+- `light.tz3210_r5afgmkl_ts0505b_2`
+- `light.tz3210_r5afgmkl_ts0505b_3`
+
+Support: color_temp (2000-6535K), xy color mode, brightness (0-254)
+
+### Vacuum Cleaner
+**Entity**: `vacuum.xiaomi_ru_1173505785_ov71gl` (Петя Петя)
+**Status**: Docked
+**Type**: Xiaomi robot vacuum with mop
+
+**Rooms** (from `sensor.xiaomi_ru_1173505785_ov71gl_room_information_p_2_16`):
+- ID 4: Спальня (Bedroom)
+- ID 3: Гостиная (Living Room)
+- ID 5: Кухня (Kitchen)
+- ID 6: Прихожая (Hallway)
+- ID 7: Ванная комната (Bathroom)
+
+**Services**:
+- `vacuum.start` — Start cleaning
+- `vacuum.pause` — Pause
+- `vacuum.stop` — Stop
+- `vacuum.return_to_base` — Dock
+- `vacuum.clean_spot` — Clean spot
+- `vacuum.set_fan_speed` — Set fan (param: `fan_speed`)
+- `vacuum.send_command` — Raw command (params: `command`, `params`)
+- Room-aware: `start_vacuum_room_sweep`, `start_zone_sweep`, `get_room_configs`, `set_room_clean_configs`
+
+**Key attributes**:
+- `sensor.xiaomi_ru_1173505785_ov71gl_room_information_p_2_16` — Room data (JSON)
+- `sensor.xiaomi_ru_1173505785_ov71gl_zone_ids_p_2_12` — Zone IDs
+- `button.xiaomi_ru_1173505785_ov71gl_auto_room_partition_a_10_5` — Auto-detect room boundaries
+
+### Water Leak Sensors
+3x HOBEIAN ZG-222Z Zigbee moisture sensors:
+- `binary_sensor.hobeian_zg_222z` — Kitchen
+- `binary_sensor.hobeian_zg_222z_2` — Bathroom
+- `binary_sensor.hobeian_zg_222z_3` — Laundry
+
+Battery sensors: `sensor.hobeian_zg_222z_battery`, `_2`, `_3`
+
+**Automations** (push to Zabbix via `rest_command`):
+- "Water Leak Alert" (`water_leak_alert`) — any sensor ON → `rest_command.zabbix_water_leak` with room name
+- "Water Leak Clear" (`water_leak_clear`) — all sensors OFF → `rest_command.zabbix_water_leak_clear`
+
+## Notes
+
+- `POST /api/states/<entity_id>` creates a virtual state representation only — it does NOT control physical devices. Use `POST /api/services/...` for actual device control.
+- Timestamp format: `YYYY-MM-DDThh:mm:ssTZD` (ISO 8601)
+- Using `?return_response` on a service that doesn't support it returns a 400 error
--- a/immich-app/backup.sh
+++ b/immich-app/backup.sh
@@ -0,0 +1,30 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+BACKUP_DIR=/mnt/backups/media
+DB_BACKUP_DIR="$BACKUP_DIR/backups"
+LOG="$BACKUP_DIR/backup.log"
+RETAIN_DAYS=14
+
+mkdir -p "$DB_BACKUP_DIR"
+
+echo "[$(date)] Starting Immich backup" >> "$LOG"
+
+# 1. Database dump (must come before file sync)
+DUMP_FILE="$DB_BACKUP_DIR/immich-db-$(date +%Y%m%dT%H%M%S).sql.gz"
+docker exec immich_postgres pg_dump --clean --if-exists \
+  --dbname=immich --username=postgres | gzip > "$DUMP_FILE"
+echo "[$(date)] DB dump: $DUMP_FILE" >> "$LOG"
+
+# 2. Rsync critical asset folders (skip thumbs and encoded-video — regeneratable)
+for DIR in library upload profile; do
+  rsync -a --delete /mnt/media/upload/$DIR/ "$BACKUP_DIR/$DIR/" >> "$LOG" 2>&1
+  echo "[$(date)] Synced $DIR" >> "$LOG"
+done
+
+# 3. Remove old DB dumps
+find "$DB_BACKUP_DIR" -name "immich-db-*.sql.gz" -mtime +$RETAIN_DAYS -delete
+echo "[$(date)] Cleaned dumps older than ${RETAIN_DAYS}d" >> "$LOG"
+
+touch "$BACKUP_DIR/.last_sync"
+echo "[$(date)] Immich backup complete" >> "$LOG"
--- a/matrix/.env
+++ b/matrix/.env
@@ -0,0 +1,7 @@
+SYNAPSE_DATA=./data/synapse
+POSTGRES_DATA=./data/postgres
+POSTGRES_USER=synapse
+POSTGRES_PASSWORD=OimW4JUSXhZBCtLHE1kFnZ7cWVbESsxynapnJ+PSw/4=
+POSTGRES_DB=synapse
+LIVEKIT_KEY=devkey
+LIVEKIT_SECRET=ef3ef4b903ca8469b09b2dd7ab6af529c4d2f3c95668f53832fc351cf67777a9
--- a/matrix/.gitignore
+++ b/matrix/.gitignore
@@ -0,0 +1 @@
+data/
--- a/matrix/README.md
+++ b/matrix/README.md
@@ -0,0 +1,105 @@
+# Matrix Home Server
+
+Self-hosted Matrix homeserver running on `mtx.alogins.net`.
+
+## Stack
+
+| Service | Purpose |
+|---------|---------|
+| Synapse | Matrix homeserver |
+| PostgreSQL | Synapse database |
+| LiveKit | MatrixRTC media server (calls) |
+| lk-jwt-service | LiveKit JWT auth for Matrix users |
+| coturn | TURN/STUN server (ICE fallback) |
+
+## Clients
+
+- **Element X** (Android/iOS) — recommended, full call support
+- **FluffyChat** — messaging only, calls not supported
+
+Connect clients to: `https://mtx.alogins.net`
+
+## Users
+
+| Username | Admin |
+|----------|-------|
+| admin | yes |
+| elizaveta | no |
+| aleksandra | no |
+
+## Managing Users
+
+```bash
+# Add user
+docker exec synapse register_new_matrix_user \
+  -c /data/homeserver.yaml \
+  -u <username> -p <password> --no-admin \
+  http://localhost:8008
+
+# Add admin
+docker exec synapse register_new_matrix_user \
+  -c /data/homeserver.yaml \
+  -u <username> -p <password> -a \
+  http://localhost:8008
+```
+
+## Start / Stop
+
+```bash
+cd /home/alvis/agap_git/matrix
+
+docker compose up -d       # start all
+docker compose down        # stop all
+docker compose restart     # restart all
+docker compose ps          # status
+docker compose logs -f     # logs
+```
+
+## Caddy
+
+Entries in `/home/alvis/agap_git/Caddyfile`:
+
+| Domain | Purpose |
+|--------|---------|
+| `mtx.alogins.net` | Synapse + well-known |
+| `lk.alogins.net` | LiveKit SFU |
+| `lkjwt.alogins.net` | LiveKit JWT service |
+
+Deploy Caddyfile changes:
+```bash
+sudo cp /home/alvis/agap_git/Caddyfile /etc/caddy/Caddyfile && sudo systemctl reload caddy
+```
+
+## Firewall Ports Required
+
+| Port | Protocol | Service |
+|------|----------|---------|
+| 443 | TCP | Caddy (HTTPS) |
+| 3478 | UDP+TCP | coturn TURN |
+| 5349 | UDP+TCP | coturn TURNS |
+| 7881 | TCP | LiveKit |
+| 49152-65535 | UDP | coturn relay |
+| 50100-50200 | UDP | LiveKit media |
+
+## Data Locations
+
+| Data | Path |
+|------|------|
+| Synapse config & media | `./data/synapse/` |
+| PostgreSQL data | `./data/postgres/` |
+| LiveKit config | `./livekit/livekit.yaml` |
+| coturn config | `./coturn/turnserver.conf` |
+
+## First-Time Setup (reference)
+
+```bash
+# Generate Synapse config
+docker run --rm \
+  -v ./data/synapse:/data \
+  -e SYNAPSE_SERVER_NAME=mtx.alogins.net \
+  -e SYNAPSE_REPORT_STATS=no \
+  matrixdotorg/synapse:latest generate
+
+# Edit database section in data/synapse/homeserver.yaml, then:
+docker compose up -d
+```
--- a/matrix/coturn/turnserver.conf
+++ b/matrix/coturn/turnserver.conf
@@ -0,0 +1,18 @@
+listening-port=3478
+tls-listening-port=5349
+
+external-ip=83.99.190.32/192.168.1.3
+
+realm=mtx.alogins.net
+server-name=mtx.alogins.net
+
+use-auth-secret
+static-auth-secret=144152cc09030796a4fd0109437dfc2089db2d5181b848d38d20c646c1d7a14b
+
+no-multicast-peers
+denied-peer-ip=10.0.0.0-10.255.255.255
+denied-peer-ip=172.16.0.0-172.31.255.255
+denied-peer-ip=192.168.0.0-192.168.255.255
+
+log-file=stdout
+no-software-attribute
--- a/matrix/docker-compose.yml
+++ b/matrix/docker-compose.yml
@@ -0,0 +1,73 @@
+services:
+  synapse:
+    image: matrixdotorg/synapse:latest
+    container_name: synapse
+    restart: unless-stopped
+    volumes:
+      - ${SYNAPSE_DATA}:/data
+      - /etc/localtime:/etc/localtime:ro
+    environment:
+      - SYNAPSE_CONFIG_PATH=/data/homeserver.yaml
+    ports:
+      - "127.0.0.1:8008:8008"
+    depends_on:
+      - db
+    networks:
+      - matrix
+      - frontend
+
+  db:
+    image: postgres:16-alpine
+    container_name: synapse-db
+    restart: unless-stopped
+    environment:
+      - POSTGRES_USER=${POSTGRES_USER}
+      - POSTGRES_PASSWORD=${POSTGRES_PASSWORD}
+      - POSTGRES_DB=${POSTGRES_DB}
+      - POSTGRES_INITDB_ARGS=--encoding=UTF-8 --lc-collate=C --lc-ctype=C
+    volumes:
+      - ${POSTGRES_DATA}:/var/lib/postgresql/data
+      - /etc/localtime:/etc/localtime:ro
+    networks:
+      - matrix
+
+  lk-jwt-service:
+    image: ghcr.io/element-hq/lk-jwt-service:latest
+    container_name: lk-jwt-service
+    restart: unless-stopped
+    ports:
+      - "127.0.0.1:8009:8080"
+    environment:
+      - LIVEKIT_JWT_BIND=:8080
+      - LIVEKIT_URL=wss://lk.alogins.net
+      - LIVEKIT_KEY=${LIVEKIT_KEY}
+      - LIVEKIT_SECRET=${LIVEKIT_SECRET}
+      - LIVEKIT_FULL_ACCESS_HOMESERVERS=mtx.alogins.net
+    extra_hosts:
+      - "mtx.alogins.net:host-gateway"
+      - "lk.alogins.net:host-gateway"
+
+  livekit:
+    image: livekit/livekit-server:latest
+    container_name: livekit
+    restart: unless-stopped
+    network_mode: host
+    volumes:
+      - ./livekit/livekit.yaml:/etc/livekit.yaml:ro
+    command: --config /etc/livekit.yaml
+
+  coturn:
+    image: coturn/coturn:latest
+    container_name: coturn
+    restart: unless-stopped
+    network_mode: host
+    volumes:
+      - ./coturn/turnserver.conf:/etc/coturn/turnserver.conf:ro
+      - /etc/localtime:/etc/localtime:ro
+
+networks:
+  matrix:
+    driver: bridge
+    internal: true
+  frontend:
+    driver: bridge
--- a/matrix/livekit/livekit.yaml
+++ b/matrix/livekit/livekit.yaml
@@ -0,0 +1,15 @@
+port: 7880
+rtc:
+  tcp_port: 7881
+  port_range_start: 50100
+  port_range_end: 50200
+  use_external_ip: true
+
+keys:
+  devkey: ef3ef4b903ca8469b09b2dd7ab6af529c4d2f3c95668f53832fc351cf67777a9
+
+room:
+  auto_create: false
+
+logging:
+  level: info
--- a/ntfy/docker-compose.yml
+++ b/ntfy/docker-compose.yml
@@ -0,0 +1,16 @@
+services:
+  ntfy:
+    image: binwiederhier/ntfy
+    container_name: ntfy
+    command: serve
+    environment:
+      - NTFY_BASE_URL=https://ntfy.alogins.net
+      - NTFY_CACHE_FILE=/var/lib/ntfy/cache.db
+      - NTFY_AUTH_FILE=/var/lib/ntfy/auth.db
+      - NTFY_AUTH_DEFAULT_ACCESS=deny-all
+      - NTFY_BEHIND_PROXY=true
+    volumes:
+      - /mnt/misc/ntfy:/var/lib/ntfy
+    ports:
+      - "8840:80"
+    restart: unless-stopped
--- a/openai/docker-compose.yml
+++ b/openai/docker-compose.yml
@@ -1,12 +1,42 @@
 services:
+  ollama:
+    image: ollama/ollama
+    container_name: ollama
+    ports:
+      - "11436:11434"
+    volumes:
+      - /mnt/ssd/ai/ollama:/root/.ollama
+      - /mnt/ssd/ai/open-webui:/app/backend/data
+    restart: always
+    environment:
+      # Allow qwen3:8b + qwen2.5:1.5b to coexist in VRAM (~6.7-7.7 GB on 8 GB GPU)
+      - OLLAMA_MAX_LOADED_MODELS=2
+      # One GPU inference at a time — prevents compute contention between models
+      - OLLAMA_NUM_PARALLEL=1
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: all
+              capabilities: [gpu]
+
+  ollama-cpu:
+    image: ollama/ollama
+    container_name: ollama-cpu
+    ports:
+      - "11435:11434"
+    volumes:
+      - /mnt/ssd/ai/ollama-cpu:/root/.ollama
+    restart: always
+   
  open-webui:
-    image: ghcr.io/open-webui/open-webui:ollama
+    image: ghcr.io/open-webui/open-webui:main
    container_name: open-webui
    ports:
      - "3125:8080"
    volumes:
-      - ollama:/root/.ollama
-      - open-webui:/app/backend/data
+      - /mnt/ssd/ai/open-webui:/app/backend/data
    restart: always
    deploy:
      resources:
@@ -18,6 +48,22 @@ services:
    environment:
      - ANTHROPIC_API_KEY=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

-volumes:
-  ollama:
-  open-webui:
+  searxng:
+    image: docker.io/searxng/searxng:latest
+    container_name: searxng
+    volumes:
+      - /mnt/ssd/ai/searxng/config/:/etc/searxng/
+      - /mnt/ssd/ai/searxng/data/:/var/cache/searxng/
+    restart: always
+    ports:
+      - "11437:8080"
+
+  qdrant:
+    image: qdrant/qdrant
+    container_name: qdrant
+    ports:
+      - "6333:6333"
+      - "6334:6334"
+    restart: always
+    volumes:
+      - /mnt/ssd/dbs/qdrant:/qdrant/storage:z
--- a/otter/docker-compose.yml
+++ b/otter/docker-compose.yml
@@ -0,0 +1,9 @@
+services:
+  otterwiki:
+    image: redimp/otterwiki:2
+    restart: unless-stopped
+    ports:
+      - 8083:80
+    volumes:
+      - /mnt/ssd/dbs/otter/app-data:/app-data
+
--- a/pihole/docker-compose.yaml
+++ b/pihole/docker-compose.yaml
@@ -0,0 +1,58 @@
+
+networks:
+  macvlan-br0:
+    driver: macvlan
+    driver_opts:
+      parent: br0
+    ipam:
+      config:
+        - subnet: 192.168.1.0/24
+          gateway: 192.168.1.1
+            # ip_range: 192.168.1.192/27
+
+services:
+  pihole:
+    container_name: pihole
+    image: pihole/pihole:latest
+      #ports:
+      # DNS Ports
+      #- "53:53/tcp"
+      #- "53:53/udp"
+      # Default HTTP Port
+      #- "80:80/tcp"
+      # Default HTTPs Port. FTL will generate a self-signed certificate
+      #- "443:443/tcp"
+      # Uncomment the below if using Pi-hole as your DHCP Server
+      #- "67:67/udp"
+      # Uncomment the line below if you are using Pi-hole as your NTP server
+      #- "123:123/udp"
+
+    dns:
+      - 8.8.8.8
+      - 1.1.1.1
+    networks:
+      macvlan-br0:
+        ipv4_address: 192.168.1.2
+    environment:
+      # Set the appropriate timezone for your location from
+      # https://en.wikipedia.org/wiki/List_of_tz_database_time_zones, e.g:
+      TZ: 'Europe/Moscow'
+      # Set a password to access the web interface. Not setting one will result in a random password being assigned
+      FTLCONF_webserver_api_password: 'correct horse 123'
+      # If using Docker's default `bridge` network setting the dns listening mode should be set to 'ALL'
+      FTLCONF_dns_listeningMode: 'ALL'
+    # Volumes store your data between container upgrades
+    volumes:
+      # For persisting Pi-hole's databases and common configuration file
+      - '/mnt/ssd/dbs/pihole:/etc/pihole'
+      # Uncomment the below if you have custom dnsmasq config files that you want to persist. Not needed for most starting fresh with Pi-hole v6. If you're upgrading from v5 you and have used this directory before, you should keep it enabled for the first v6 container start to allow for a complete migration. It can be removed afterwards. Needs environment variable FTLCONF_misc_etc_dnsmasq_d: 'true'
+      #- './etc-dnsmasq.d:/etc/dnsmasq.d'
+    cap_add:
+      # See https://github.com/pi-hole/docker-pi-hole#note-on-capabilities
+      # Required if you are using Pi-hole as your DHCP server, else not needed
+      - NET_ADMIN
+      # Required if you are using Pi-hole as your NTP client to be able to set the host's system time
+      - SYS_TIME
+      # Optional, if Pi-hole should get some more processing time
+      - SYS_NICE
+    restart: unless-stopped
--- a/seafile/backup.sh
+++ b/seafile/backup.sh
@@ -0,0 +1,44 @@
+#!/bin/bash
+# Seafile backup script.
+# Backs up MySQL databases and seafile data directory.
+# Runs every 3 days via root crontab. Keeps last 5 backups.
+# Notifies Zabbix (item seafile.backup.ts, id 70369 on AgapHost) after success.
+
+set -euo pipefail
+
+BACKUP_DIR="/mnt/backups/seafile"
+DATA_DIR="/mnt/misc/seafile"
+DATE=$(date '+%Y%m%d-%H%M')
+DEST="$BACKUP_DIR/$DATE"
+
+mkdir -p "$DEST"
+
+# Dump all three Seafile databases from the running container
+for DB in ccnet_db seafile_db seahub_db; do
+  docker exec seafile-mysql mysqldump \
+    -u seafile -pFWsYYeZa15ro6x \
+    --single-transaction "$DB" > "$DEST/${DB}.sql"
+  echo "Dumped: $DB"
+done
+
+# Copy seafile data (libraries, config — excludes mysql and caddy dirs)
+rsync -a --delete \
+  --exclude='seafile-mysql/' \
+  --exclude='seafile-caddy/' \
+  "$DATA_DIR/" "$DEST/data/"
+
+echo "$(date): Backup complete: $DEST"
+ls "$DEST/"
+
+# Notify Zabbix
+if [[ -f /root/.zabbix_token ]]; then
+  ZABBIX_TOKEN=$(cat /root/.zabbix_token)
+  curl -s -X POST http://localhost:81/api_jsonrpc.php \
+    -H "Content-Type: application/json" \
+    -H "Authorization: Bearer $ZABBIX_TOKEN" \
+    -d "{\"jsonrpc\":\"2.0\",\"method\":\"history.push\",\"id\":1,\"params\":{\"itemid\":\"70369\",\"value\":\"$(date '+%Y-%m-%d %H:%M')\"}}" > /dev/null \
+    && echo "Zabbix notified."
+fi
+
+# Rotate: keep last 5 backups
+ls -1dt "$BACKUP_DIR"/[0-9]*-[0-9]* 2>/dev/null | tail -n +6 | xargs -r rm -rf
--- a/seafile/caddy.yml
+++ b/seafile/caddy.yml
@@ -0,0 +1,26 @@
+services:
+
+  caddy:
+    image: ${SEAFILE_CADDY_IMAGE:-lucaslorentz/caddy-docker-proxy:2.9-alpine}
+    restart: unless-stopped
+    container_name: seafile-caddy
+    ports:
+      - 8077:80
+      - 4433:443
+    environment:
+      - CADDY_INGRESS_NETWORKS=seafile-net
+    volumes:
+      - /var/run/docker.sock:/var/run/docker.sock
+      - ${SEAFILE_CADDY_VOLUME:-/opt/seafile-caddy}:/data/caddy
+    networks:
+      - seafile-net
+    healthcheck:
+      test: ["CMD-SHELL", "curl --fail http://localhost:2019/metrics || exit 1"]
+      start_period: 20s
+      interval: 20s
+      timeout: 5s
+      retries: 3
+
+networks:
+  seafile-net:
+    name: seafile-net
--- a/seafile/onlyoffice.yml
+++ b/seafile/onlyoffice.yml
@@ -0,0 +1,20 @@
+services:
+  onlyoffice:
+    image: ${ONLYOFFICE_IMAGE:-onlyoffice/documentserver:8.1.0.1}
+    container_name: seafile-onlyoffice
+    restart: unless-stopped
+    environment:
+      - JWT_ENABLED=true
+      - JWT_SECRET=${ONLYOFFICE_JWT_SECRET:?Variable is not set or empty}
+    volumes:
+      - "${ONLYOFFICE_VOLUME:-/opt/onlyoffice}:/var/lib/onlyoffice"
+    ports:
+      - "127.0.0.1:6233:80"
+    extra_hosts:
+      - "docs.alogins.net:host-gateway"
+    networks:
+      - seafile-net
+
+networks:
+  seafile-net:
+    name: seafile-net
--- a/seafile/seadoc.yml
+++ b/seafile/seadoc.yml
@@ -0,0 +1,40 @@
+services:
+
+  seadoc:
+    image: ${SEADOC_IMAGE:-seafileltd/sdoc-server:2.0-latest}
+    container_name: seadoc
+    restart: unless-stopped
+    volumes:
+      - ${SEADOC_VOLUME:-/opt/seadoc-data/}:/shared
+    # ports:
+    #   - "80:80"
+    environment:
+      - DB_HOST=${SEAFILE_MYSQL_DB_HOST:-db}
+      - DB_PORT=${SEAFILE_MYSQL_DB_PORT:-3306}
+      - DB_USER=${SEAFILE_MYSQL_DB_USER:-seafile}
+      - DB_PASSWORD=${SEAFILE_MYSQL_DB_PASSWORD:?Variable is not set or empty}
+      - DB_NAME=${SEADOC_MYSQL_DB_NAME:-${SEAFILE_MYSQL_DB_SEAHUB_DB_NAME:-seahub_db}}
+      - TIME_ZONE=${TIME_ZONE:-Etc/UTC}
+      - JWT_PRIVATE_KEY=${JWT_PRIVATE_KEY:?Variable is not set or empty}
+      - NON_ROOT=${NON_ROOT:-false}
+      - SEAHUB_SERVICE_URL=${SEAFILE_SERVICE_URL:-http://seafile}
+    labels:
+      caddy: ${SEAFILE_SERVER_PROTOCOL:-http}://${SEAFILE_SERVER_HOSTNAME:?Variable is not set or empty}
+      caddy.@ws.0_header: "Connection *Upgrade*"
+      caddy.@ws.1_header: "Upgrade websocket"
+      caddy.0_reverse_proxy: "@ws {{upstreams 80}}"
+      caddy.1_handle_path: "/socket.io/*"
+      caddy.1_handle_path.0_rewrite: "* /socket.io{uri}"
+      caddy.1_handle_path.1_reverse_proxy: "{{upstreams 80}}"
+      caddy.2_handle_path: "/sdoc-server/*"
+      caddy.2_handle_path.0_rewrite: "* {uri}"
+      caddy.2_handle_path.1_reverse_proxy: "{{upstreams 80}}"
+    depends_on:
+      db:
+        condition: service_healthy
+    networks:
+      - seafile-net
+
+networks:
+  seafile-net:
+    name: seafile-net
--- a/seafile/seafile-server.yml
+++ b/seafile/seafile-server.yml
@@ -0,0 +1,103 @@
+services:
+  db:
+    image: ${SEAFILE_DB_IMAGE:-mariadb:10.11}
+    container_name: seafile-mysql
+    restart: unless-stopped
+    environment:
+      - MYSQL_ROOT_PASSWORD=${INIT_SEAFILE_MYSQL_ROOT_PASSWORD:-}
+      - MYSQL_LOG_CONSOLE=true
+      - MARIADB_AUTO_UPGRADE=1
+    volumes:
+      - "${SEAFILE_MYSQL_VOLUME:-/opt/seafile-mysql/db}:/var/lib/mysql"
+    networks:
+      - seafile-net
+    healthcheck:
+      test:
+        [
+          "CMD",
+          "/usr/local/bin/healthcheck.sh",
+          "--connect",
+          "--mariadbupgrade",
+          "--innodb_initialized",
+        ]
+      interval: 20s
+      start_period: 30s
+      timeout: 5s
+      retries: 10
+
+  redis:
+    image: ${SEAFILE_REDIS_IMAGE:-redis}
+    container_name: seafile-redis
+    restart: unless-stopped
+    command:
+      - /bin/sh
+      - -c
+      - redis-server --requirepass "$$REDIS_PASSWORD"
+    environment:
+      - REDIS_PASSWORD=${REDIS_PASSWORD:-}
+    networks:
+      - seafile-net
+
+  seafile:
+    image: ${SEAFILE_IMAGE:-seafileltd/seafile-mc:13.0-latest}
+    container_name: seafile
+    restart: unless-stopped
+    ports:
+      - "127.0.0.1:8078:80"
+    volumes:
+      - ${SEAFILE_VOLUME:-/opt/seafile-data}:/shared
+    environment:
+      - SEAFILE_MYSQL_DB_HOST=${SEAFILE_MYSQL_DB_HOST:-db}
+      - SEAFILE_MYSQL_DB_PORT=${SEAFILE_MYSQL_DB_PORT:-3306}
+      - SEAFILE_MYSQL_DB_USER=${SEAFILE_MYSQL_DB_USER:-seafile}
+      - SEAFILE_MYSQL_DB_PASSWORD=${SEAFILE_MYSQL_DB_PASSWORD:?Variable is not set or empty}
+      - INIT_SEAFILE_MYSQL_ROOT_PASSWORD=${INIT_SEAFILE_MYSQL_ROOT_PASSWORD:-}
+      - SEAFILE_MYSQL_DB_CCNET_DB_NAME=${SEAFILE_MYSQL_DB_CCNET_DB_NAME:-ccnet_db}
+      - SEAFILE_MYSQL_DB_SEAFILE_DB_NAME=${SEAFILE_MYSQL_DB_SEAFILE_DB_NAME:-seafile_db}
+      - SEAFILE_MYSQL_DB_SEAHUB_DB_NAME=${SEAFILE_MYSQL_DB_SEAHUB_DB_NAME:-seahub_db}
+      - TIME_ZONE=${TIME_ZONE:-Etc/UTC}
+      - INIT_SEAFILE_ADMIN_EMAIL=${INIT_SEAFILE_ADMIN_EMAIL:-me@example.com}
+      - INIT_SEAFILE_ADMIN_PASSWORD=${INIT_SEAFILE_ADMIN_PASSWORD:-asecret}
+      - SEAFILE_SERVER_HOSTNAME=${SEAFILE_SERVER_HOSTNAME:?Variable is not set or empty}
+      - SEAFILE_SERVER_PROTOCOL=${SEAFILE_SERVER_PROTOCOL:-http}
+      - SITE_ROOT=${SITE_ROOT:-/}
+      - NON_ROOT=${NON_ROOT:-false}
+      - JWT_PRIVATE_KEY=${JWT_PRIVATE_KEY:?Variable is not set or empty}
+      - SEAFILE_LOG_TO_STDOUT=${SEAFILE_LOG_TO_STDOUT:-false}
+      - ENABLE_GO_FILESERVER=${ENABLE_GO_FILESERVER:-true}
+      - ENABLE_SEADOC=${ENABLE_SEADOC:-true}
+      - SEADOC_SERVER_URL=${SEAFILE_SERVER_PROTOCOL:-http}://${SEAFILE_SERVER_HOSTNAME:?Variable is not set or empty}/sdoc-server
+      - CACHE_PROVIDER=${CACHE_PROVIDER:-redis}
+      - REDIS_HOST=${REDIS_HOST:-redis}
+      - REDIS_PORT=${REDIS_PORT:-6379}
+      - REDIS_PASSWORD=${REDIS_PASSWORD:-}
+      - MEMCACHED_HOST=${MEMCACHED_HOST:-memcached}
+      - MEMCACHED_PORT=${MEMCACHED_PORT:-11211}
+      - ENABLE_NOTIFICATION_SERVER=${ENABLE_NOTIFICATION_SERVER:-false}
+      - INNER_NOTIFICATION_SERVER_URL=${INNER_NOTIFICATION_SERVER_URL:-http://notification-server:8083}
+      - NOTIFICATION_SERVER_URL=${NOTIFICATION_SERVER_URL:-${SEAFILE_SERVER_PROTOCOL:-http}://${SEAFILE_SERVER_HOSTNAME:?Variable is not set or empty}/notification}
+      - ENABLE_SEAFILE_AI=${ENABLE_SEAFILE_AI:-false}
+      - ENABLE_FACE_RECOGNITION=${ENABLE_FACE_RECOGNITION:-false}
+      - SEAFILE_AI_SERVER_URL=${SEAFILE_AI_SERVER_URL:-http://seafile-ai:8888}
+      - SEAFILE_AI_SECRET_KEY=${JWT_PRIVATE_KEY:?Variable is not set or empty}
+      - MD_FILE_COUNT_LIMIT=${MD_FILE_COUNT_LIMIT:-100000}
+    labels:
+      caddy: ${SEAFILE_SERVER_PROTOCOL:-http}://${SEAFILE_SERVER_HOSTNAME:?Variable is not set or empty}
+      caddy.reverse_proxy: "{{upstreams 80}}"
+    healthcheck:
+      test: ["CMD-SHELL", "curl -f http://localhost:80 || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 10s
+    depends_on:
+      db:
+        condition: service_healthy
+      redis:
+        condition: service_started
+    networks:
+      - seafile-net
+
+networks:
+  seafile-net:
+    name: seafile-net
--- a/users-backup.sh
+++ b/users-backup.sh
@@ -0,0 +1,25 @@
+#!/bin/bash
+# Backup /mnt/misc/alvis and /mnt/misc/liza to /mnt/backups/users/
+# Runs every 3 days via root crontab.
+# Notifies Zabbix (item users.backup.ts, id 70379 on AgapHost) after success.
+
+set -euo pipefail
+
+DEST=/mnt/backups/users
+
+mkdir -p "$DEST/alvis" "$DEST/liza"
+
+rsync -a --delete /mnt/misc/alvis/ "$DEST/alvis/"
+rsync -a --delete /mnt/misc/liza/ "$DEST/liza/"
+
+echo "$(date): Backup complete."
+
+# Notify Zabbix (token stored in /root/.zabbix_token)
+if [[ -f /root/.zabbix_token ]]; then
+  ZABBIX_TOKEN=$(cat /root/.zabbix_token)
+  curl -s -X POST http://localhost:81/api_jsonrpc.php \
+    -H "Content-Type: application/json" \
+    -H "Authorization: Bearer $ZABBIX_TOKEN" \
+    -d "{\"jsonrpc\":\"2.0\",\"method\":\"history.push\",\"id\":1,\"params\":{\"itemid\":\"70379\",\"value\":\"$(date '+%Y-%m-%d %H:%M')\"}}" > /dev/null \
+    && echo "Zabbix notified."
+fi
--- a/vaultwarden/backup.sh
+++ b/vaultwarden/backup.sh
@@ -0,0 +1,41 @@
+#!/bin/bash
+# Vaultwarden backup — uses built-in container backup command (safe with live DB).
+# Runs every 3 days via root crontab. Keeps last 5 backups.
+# Notifies Zabbix (item vaultwarden.backup.ts, id 70368 on AgapHost) after success.
+
+set -euo pipefail
+
+BACKUP_DIR="/mnt/backups/vaultwarden"
+DATA_DIR="/mnt/ssd/dbs/vw-data"
+DATE=$(date '+%Y%m%d-%H%M')
+DEST="$BACKUP_DIR/$DATE"
+
+mkdir -p "$DEST"
+
+# Run built-in backup inside container — writes db_<timestamp>.sqlite3 to /data/ on the host
+docker exec vaultwarden /vaultwarden backup 2>&1
+
+# Move the newly created sqlite3 backup file out of the data dir
+find "$DATA_DIR" -maxdepth 1 -name 'db_*.sqlite3' -newer "$DATA_DIR/db.sqlite3" | xargs -r mv -t "$DEST/"
+
+# Copy config and RSA keys
+cp "$DATA_DIR/config.json" "$DEST/"
+cp "$DATA_DIR"/rsa_key* "$DEST/"
+[ -d "$DATA_DIR/attachments" ] && cp -r "$DATA_DIR/attachments" "$DEST/"
+[ -d "$DATA_DIR/sends" ]       && cp -r "$DATA_DIR/sends"       "$DEST/"
+
+echo "$(date): Backup complete: $DEST"
+ls "$DEST/"
+
+# Notify Zabbix (token stored in /root/.zabbix_token)
+if [[ -f /root/.zabbix_token ]]; then
+  ZABBIX_TOKEN=$(cat /root/.zabbix_token)
+  curl -s -X POST http://localhost:81/api_jsonrpc.php \
+    -H "Content-Type: application/json" \
+    -H "Authorization: Bearer $ZABBIX_TOKEN" \
+    -d "{\"jsonrpc\":\"2.0\",\"method\":\"history.push\",\"id\":1,\"params\":{\"itemid\":\"70368\",\"value\":\"$(date '+%Y-%m-%d %H:%M')\"}}" > /dev/null \
+    && echo "Zabbix notified."
+fi
+
+# Rotate: keep last 5 backups
+ls -1dt "$BACKUP_DIR"/[0-9]*-[0-9]* 2>/dev/null | tail -n +6 | xargs -r rm -rf
--- a/vaultwarden/docker-compose.yml
+++ b/vaultwarden/docker-compose.yml
@@ -0,0 +1,12 @@
+services:
+  vaultwarden:
+    image: vaultwarden/server:latest
+    container_name: vaultwarden
+    restart: unless-stopped
+    environment:
+      DOMAIN: "https://vw.alogins.net"
+      ADMIN_TOKEN: $$argon2id$$v=19$$m=65540,t=3,p=4$$bkE5Y1grLzF4czZiUk9tcWR6WTlGNC9CQmxGeHg0R1JUMFBrY2l0SVZocz0$$hn0snCmQkzDTEBzPYGQxFNmHxTgpxQ+O8OvzOhR3/a0
+    volumes:
+      - /mnt/ssd/dbs/vw-data/:/data/
+    ports:
+      - 127.0.0.1:8041:80
Author	SHA1	Message	Date
Alvis	e04f9059ae	Add Matrix homeserver with MatrixRTC calling support - Synapse + PostgreSQL + coturn + LiveKit + lk-jwt-service - Caddy entries for mtx.alogins.net, lk.alogins.net, lkjwt.alogins.net - well-known endpoints for Matrix client/server discovery and RTC transport - Users: admin, elizaveta, aleksandra Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-15 14:12:13 +00:00
Alvis	002f9863b0	Add users backup script with Zabbix notification Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-09 06:26:32 +00:00
Alvis	77c7cd09aa	Update CLAUDE.md: expand Seafile wiki page description	2026-03-08 16:12:13 +00:00
Alvis	b66a74df06	Add Seafile backup script with Zabbix monitoring - backup.sh: mysqldump all 3 DBs + rsync seafile-data, runs every 3 days via root crontab, keeps last 5 backups in /mnt/backups/seafile - Notifies Zabbix trapper item seafile.backup.ts (id 70369) on AgapHost Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-08 16:06:14 +00:00
Alvis	b8db06cd21	Fix OnlyOffice→Seafile connectivity (hairpin NAT) Add extra_hosts: docs.alogins.net:host-gateway so OnlyOffice container can reach Seafile's callback URL without going through the public IP. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-08 15:45:08 +00:00
Alvis	7e889d8530	Add OnlyOffice integration for Seafile - seafile/onlyoffice.yml: OnlyOffice Document Server 8.1 with JWT auth - Expose on 127.0.0.1:6233, proxied via Caddy at office.alogins.net - Caddyfile: add office.alogins.net → localhost:6233 - JWT secret stored in Vaultwarden (ONLYOFFICE_JWT_SECRET) - seahub_settings.py configured inside container with ENABLE_ONLYOFFICE=True Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-08 15:36:30 +00:00
Alvis	73ba559593	Update CLAUDE.md: add Seafile wiki page	2026-03-08 15:18:51 +00:00
Alvis	10cb24b7e5	Add Seafile service and update Caddyfile - seafile/: docker compose setup (seafile-mc 13, mariadb, redis, seadoc, caddy-proxy) - Expose seafile on 127.0.0.1:8078, proxied via Caddy at docs.alogins.net - Fix: SEAFILE_SERVER_PROTOCOL=https to avoid CSRF errors - Fix: TIME_ZONE=Asia/Dubai (Etc/UTC+4 was invalid) - Caddyfile: add docs.alogins.net → localhost:8078 - .gitignore: exclude seafile/.env (credentials stored in Vaultwarden) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-08 15:11:08 +00:00
Alvis	20c318b3c1	Update CLAUDE.md: add Vaultwarden service and wiki page	2026-03-08 13:45:34 +00:00
Alvis	8873e441c2	Add Vaultwarden backup script with Zabbix monitoring - backup.sh: runs every 3 days via root crontab, uses built-in container backup command, copies db/config/rsa_key to /mnt/backups/vaultwarden, keeps last 5 backups, notifies Zabbix item vaultwarden.backup.ts (id 70368) - Zabbix trigger fires if no backup received in 4 days Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-08 13:44:11 +00:00
Alvis	d72fd95dfd	Add Vaultwarden service and update Caddyfile - Add vaultwarden/docker-compose.yml (port 8041, data on /mnt/ssd/dbs/vw-data) - Update Caddyfile with all current services including vw.alogins.net Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-08 13:13:49 +00:00
Alvis	87eb4fb765	Remove adolf — moved to separate repo (alvis/adolf)	2026-03-08 07:06:07 +00:00
Alvis	e2e15009e2	Add Immich backup script Daily backup at 02:30 via root cron: DB dump + rsync of library/upload/profile to /mnt/backups/media/. Retains 14 days of DB dumps. Monitored via Zabbix immich.backup.age item with High trigger if stale >25h. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-07 18:28:58 +00:00
Alvis	5017827af2	cleaning	2026-03-07 17:50:46 +00:00
Alvis	a30936f120	wiki search people tested pipeline	2026-03-05 11:22:34 +00:00