Initial release: Multi-provider AI chat with RAG

FastAPI backend (wiki-vector-chat.py) with Odysseus-style frontend.
Features: multi-provider LLM, Wiki KB + VectorDB RAG, session history,
chat modes, save-to-wiki, markdown rendering, SSE streaming.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

wiki-chat-server.py

@@ -0,0 +1,90 @@
+#!/usr/bin/env python3
+"""Wiki Chat LLM Proxy Server on port 8098 using Flask"""
+
+import json
+import os
+import urllib.request
+from flask import Flask, request, jsonify, make_response
+
+app = Flask(__name__)
+
+ZAI_TOKEN = os.environ.get('ZAI_API_TOKEN', '')
+
+@app.after_request
+def add_cors(response):
+    h = 'Access-Control-Allow-Origin'
+    response.headers[h] = '*'
+    response.headers['Access-Control-Allow-Methods'] = 'POST, OPTIONS'
+    response.headers['Access-Control-Allow-Headers'] = 'Content-Type, Authorization'
+    return response
+
+def call_zai(messages, token=None):
+    api_url = 'https://api.z.ai/api/coding/paas/v4/chat/completions'
+    payload = json.dumps({
+        'model': 'glm-4-plus',
+        'messages': messages,
+        'temperature': 0.7,
+        'max_tokens': 2000,
+    }).encode()
+    headers = {'Content-Type': 'application/json'}
+    if token:
+        headers['Authorization'] = 'Bearer ' + token
+    req = urllib.request.Request(api_url, data=payload, headers=headers, method='POST')
+    with urllib.request.urlopen(req, timeout=60) as resp:
+        data = json.loads(resp.read().decode())
+        return data.get('choices', [{}])[0].get('message', {}).get('content', '')
+
+@app.route('/chat/wiki', methods=['POST', 'OPTIONS'])
+def chat_wiki():
+    if request.method == 'OPTIONS':
+        return make_response('', 200)
+    try:
+        body = request.get_json(force=True)
+        messages = body.get('messages', [])
+        if not messages:
+            return jsonify({'error': 'messages required'}), 400
+        token = body.get('token') or ZAI_TOKEN
+        if not token:
+            return jsonify({'error': 'No token provided'}), 401
+        content = call_zai(messages, token)
+        return jsonify({'response': content})
+    except urllib.error.HTTPError as e:
+        err_body = e.read().decode() if e.fp else ''
+        try:
+            err_json = json.loads(err_body)
+            err_msg = err_json.get('error', {}).get('message', err_json.get('message', str(e)))
+        except Exception:
+            err_msg = str(e)
+        return jsonify({'error': err_msg}), e.code
+    except Exception as e:
+        return jsonify({'error': str(e)}), 500
+
+@app.route('/chat/wiki-tunnel', methods=['POST', 'OPTIONS'])
+def chat_wiki_tunnel():
+    if request.method == 'OPTIONS':
+        return make_response('', 200)
+    try:
+        body = request.get_json(force=True)
+        messages = body.get('messages', [])
+        if not messages:
+            return jsonify({'error': 'messages required'}), 400
+        token = ZAI_TOKEN
+        if not token:
+            return jsonify({'error': 'No server token configured. Use Token mode and paste your API key.'}), 503
+        content = call_zai(messages, token)
+        return jsonify({'response': content})
+    except urllib.error.HTTPError as e:
+        err_body = e.read().decode() if e.fp else ''
+        try:
+            err_json = json.loads(err_body)
+            err_msg = err_json.get('error', {}).get('message', err_json.get('message', str(e)))
+        except Exception:
+            err_msg = str(e)
+        return jsonify({'error': err_msg}), e.code
+    except Exception as e:
+        return jsonify({'error': str(e)}), 500
+
+if __name__ == '__main__':
+    port = int(os.environ.get('WIKI_CHAT_PORT', 8098))
+    print(f'[WikiChat] LLM proxy running on port {port}')
+    app.run(host='0.0.0.0', port=port)