refactor: split openai/ into ollama/, openwebui/, searxng/

Move services out of the monolithic openai/docker-compose.yml:
- ollama/ — ollama GPU + CPU inference servers
- openwebui/ — open-webui chat UI (uses env var for ANTHROPIC_API_KEY)
- searxng/ — SearXNG container + searxng-mcp MCP server (port 3102)

openai/ now contains only: litellm, langfuse, qdrant, faster-whisper,
silero-tts, pipecat.

searxng-mcp exposes a single searxng_search tool via MCP HTTP on :3102.
Registered in ~/.claude.json as the "searxng" MCP server.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

ollama/docker-compose.yml

@@ -0,0 +1,27 @@
+services:
+  ollama:
+    image: ollama/ollama
+    container_name: ollama
+    ports:
+      - "11436:11434"
+    volumes:
+      - /mnt/ssd/ai/ollama:/root/.ollama
+    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
+      # Force all layers to GPU — fail instead of falling back to CPU
+      - OLLAMA_NUM_GPU=999
+    runtime: nvidia
+    mem_limit: 4g
+
+  ollama-cpu:
+    image: ollama/ollama
+    container_name: ollama-cpu
+    ports:
+      - "11435:11434"
+    volumes:
+      - /mnt/ssd/ai/ollama-cpu:/root/.ollama
+    restart: always

openai/docker-compose.yml

@@ -1,65 +1,4 @@
 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
-      # Force all layers to GPU — fail instead of falling back to CPU
-      - OLLAMA_NUM_GPU=999
-    runtime: nvidia
-    mem_limit: 4g
-
-  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:main
-    container_name: open-webui
-    ports:
-      - "3125:8080"
-    volumes:
-      - /mnt/ssd/ai/open-webui:/app/backend/data
-    extra_hosts:
-      - "host.docker.internal:host-gateway"
-    restart: always
-    deploy:
-      resources:
-        reservations:
-          devices:
-            - driver: nvidia
-              count: all
-              capabilities: [gpu]
-    environment:
-      - ANTHROPIC_API_KEY=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
-      - OPENAI_API_BASE_URL=http://host.docker.internal:4000/v1
-      - OPENAI_API_KEY=dummy
-      # STT — Faster-Whisper large-v3-turbo
-      - AUDIO_STT_ENGINE=openai
-      - AUDIO_STT_OPENAI_API_BASE_URL=http://host.docker.internal:8880/v1
-      - AUDIO_STT_OPENAI_API_KEY=dummy
-      - AUDIO_STT_MODEL=deepdml/faster-whisper-large-v3-turbo-ct2
-      # TTS — Silero v4
-      - AUDIO_TTS_ENGINE=openai
-      - AUDIO_TTS_OPENAI_API_BASE_URL=http://host.docker.internal:8881/v1
-      - AUDIO_TTS_OPENAI_API_KEY=dummy
-      - AUDIO_TTS_MODEL=silero
-      - AUDIO_TTS_VOICE=onyx
-
   litellm-db:
     image: postgres:16-alpine
     container_name: litellm-db
@@ -119,16 +58,6 @@ services:
       - langfuse-db
     restart: always
 
-  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

openwebui/docker-compose.yml

@@ -0,0 +1,33 @@
+services:
+  open-webui:
+    image: ghcr.io/open-webui/open-webui:main
+    container_name: open-webui
+    ports:
+      - "3125:8080"
+    volumes:
+      - /mnt/ssd/ai/open-webui:/app/backend/data
+    extra_hosts:
+      - "host.docker.internal:host-gateway"
+    restart: always
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: all
+              capabilities: [gpu]
+    environment:
+      - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY}
+      - OPENAI_API_BASE_URL=http://host.docker.internal:4000/v1
+      - OPENAI_API_KEY=dummy
+      # STT — Faster-Whisper large-v3-turbo
+      - AUDIO_STT_ENGINE=openai
+      - AUDIO_STT_OPENAI_API_BASE_URL=http://host.docker.internal:8880/v1
+      - AUDIO_STT_OPENAI_API_KEY=dummy
+      - AUDIO_STT_MODEL=deepdml/faster-whisper-large-v3-turbo-ct2
+      # TTS — Silero v4
+      - AUDIO_TTS_ENGINE=openai
+      - AUDIO_TTS_OPENAI_API_BASE_URL=http://host.docker.internal:8881/v1
+      - AUDIO_TTS_OPENAI_API_KEY=dummy
+      - AUDIO_TTS_MODEL=silero
+      - AUDIO_TTS_VOICE=onyx

searxng/docker-compose.yml

@@ -0,0 +1,19 @@
+services:
+  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"
+
+  searxng-mcp:
+    build: ./mcp
+    container_name: searxng-mcp
+    network_mode: host
+    restart: unless-stopped
+    environment:
+      - PORT=3102
+      - SEARXNG_URL=http://localhost:11437

searxng/mcp/Dockerfile

@@ -0,0 +1,6 @@
+FROM node:22-slim
+WORKDIR /app
+COPY package.json ./
+RUN npm install --omit=dev
+COPY server.js ./
+CMD ["node", "server.js"]

searxng/mcp/package.json

@@ -0,0 +1,11 @@
+{
+  "name": "searxng-mcp",
+  "version": "1.0.0",
+  "type": "module",
+  "main": "server.js",
+  "dependencies": {
+    "@modelcontextprotocol/sdk": "^1.12.0",
+    "express": "^4.19.0",
+    "zod": "^3.23.0"
+  }
+}

searxng/mcp/server.js

@@ -0,0 +1,98 @@
+import express from 'express';
+import { McpServer } from '@modelcontextprotocol/sdk/server/mcp.js';
+import { StreamableHTTPServerTransport } from '@modelcontextprotocol/sdk/server/streamableHttp.js';
+import { SSEServerTransport } from '@modelcontextprotocol/sdk/server/sse.js';
+import { z } from 'zod';
+
+const PORT = parseInt(process.env.PORT || '3102');
+const SEARXNG_URL = (process.env.SEARXNG_URL || 'http://localhost:11437').replace(/\/$/, '');
+
+function createServer() {
+  const server = new McpServer({ name: 'searxng-mcp', version: '1.0.0' });
+
+  server.tool(
+    'searxng_search',
+    'Search the web using the self-hosted SearXNG meta-search engine. Returns titles, URLs, and content snippets from multiple search engines.',
+    {
+      query: z.string().describe('Search query'),
+      categories: z.enum(['general', 'news', 'images', 'videos', 'science', 'files', 'social_media', 'it'])
+        .optional()
+        .describe('Search category, default: general'),
+      language: z.string().optional().describe('Language code (e.g. "ru", "en", "auto"), default: auto'),
+      time_range: z.enum(['day', 'week', 'month', 'year']).optional().describe('Limit results to time range'),
+      limit: z.number().optional().describe('Max results to return, default 10'),
+    },
+    async ({ query, categories, language, time_range, limit = 10 }) => {
+      try {
+        const params = new URLSearchParams({
+          q: query,
+          format: 'json',
+          categories: categories || 'general',
+          language: language || 'auto',
+        });
+        if (time_range) params.set('time_range', time_range);
+
+        const res = await fetch(`${SEARXNG_URL}/search?${params}`);
+        if (!res.ok) {
+          return { content: [{ type: 'text', text: `SearXNG returned HTTP ${res.status}` }], isError: true };
+        }
+
+        const data = await res.json();
+        const results = (data.results || []).slice(0, limit).map(r => ({
+          title: r.title || null,
+          url: r.url || null,
+          content: (r.content || '').slice(0, 500) || null,
+          engine: r.engine || null,
+          score: r.score != null ? Math.round(r.score * 100) / 100 : null,
+          publishedDate: r.publishedDate || null,
+        }));
+
+        const out = {
+          query: data.query,
+          totalResults: data.number_of_results,
+          count: results.length,
+          results,
+        };
+
+        return { content: [{ type: 'text', text: JSON.stringify(out, null, 2) }] };
+      } catch (e) {
+        return { content: [{ type: 'text', text: `Error: ${e.message}` }], isError: true };
+      }
+    }
+  );
+
+  return server;
+}
+
+const app = express();
+app.use(express.json());
+
+const sseTransports = new Map();
+
+app.all('/mcp', async (req, res) => {
+  try {
+    const transport = new StreamableHTTPServerTransport({ sessionIdGenerator: undefined });
+    res.on('close', () => transport.close());
+    await createServer().connect(transport);
+    await transport.handleRequest(req, res, req.body);
+  } catch (e) {
+    if (!res.headersSent) res.status(500).json({ error: e.message });
+  }
+});
+
+app.get('/sse', async (req, res) => {
+  const transport = new SSEServerTransport('/messages', res);
+  sseTransports.set(transport.sessionId, transport);
+  res.on('close', () => sseTransports.delete(transport.sessionId));
+  await createServer().connect(transport);
+});
+
+app.post('/messages', async (req, res) => {
+  const transport = sseTransports.get(req.query.sessionId);
+  if (!transport) return res.status(400).send('Unknown session');
+  await transport.handlePostMessage(req, res);
+});
+
+app.get('/health', (_, res) => res.json({ status: 'ok', searxng: SEARXNG_URL }));
+
+app.listen(PORT, () => console.log(`searxng-mcp listening on :${PORT}`));