From 776546f31872f17286c54f552ad26dcdf468c413 Mon Sep 17 00:00:00 2001
From: fyah <fyah@nerd-brain.local>
Date: Mon, 27 Apr 2026 10:04:25 +0800
Subject: [PATCH] =?UTF-8?q?feat:=20=E5=88=87=E6=8D=A2=E8=87=B3=E5=BC=A0?=
 =?UTF-8?q?=E4=B8=89=20(Qwen=203.6-35B-A3B)=20=E6=A8=A1=E5=9E=8B?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- 新增 zhangsan provider 配置
- anchor_writer.py 优先使用 zhangsan
- 测试连通性成功，生成笔记符合新规范
---
 anchor_writer.py | 26 ++++++++++++++++++--------
 1 file changed, 18 insertions(+), 8 deletions(-)

diff --git a/anchor_writer.py b/anchor_writer.py
index 392d843..c498398 100644
--- a/anchor_writer.py
+++ b/anchor_writer.py
@@ -24,24 +24,34 @@ def get_api_config():
     if not providers:
         raise ValueError("No custom providers found in config")
     
-    # Find bailian provider
+    # Find zhangsan provider first, then bailian
+    zhangsan = None
     bailian = None
     for p in providers:
-        if p.get('name') == 'bailian':
+        if p.get('name') == 'zhangsan':
+            zhangsan = p
+        elif p.get('name') == 'bailian':
             bailian = p
-            break
             
-    if not bailian:
-        bailian = providers[0] # Fallback to first
-        
-    return bailian['api_key'], bailian['base_url'], bailian.get('model', 'qwen3.5-plus')
+    if zhangsan:
+        return zhangsan['api_key'], zhangsan['base_url'], zhangsan.get('model', 'Qwen3.6-35B-A3B')
+    elif bailian:
+        return bailian['api_key'], bailian['base_url'], bailian.get('model', 'qwen3.5-plus')
+    else:
+        return providers[0]['api_key'], providers[0]['base_url'], providers[0].get('model', 'default')
 
 def call_model(content):
     """Call Qwen API to generate summary, tags, and emotion scores"""
     api_key, base_url, model = get_api_config()
     
     # Simpler prompt, no response_format to avoid JSON parsing issues
-    prompt = f"Analyze this text. Output ONLY a valid JSON object. No markdown, no explanation. Keys: summary (max 50 words), tags (list of 3 keywords), valence (-1.0 to 1.0), arousal (0.0 to 1.0). Text: {content}"
+    prompt = f"""Analyze this text. Output ONLY a valid JSON object. No markdown, no explanation. 
+Keys: 
+- summary (max 50 words, append URLs if external resources/GitHub mentioned)
+- tags (list of EXACTLY 2 strings, format '#Category/ProjectName', e.g., ['#memory/nerd-brain', '#llm/local-deploy'])
+- valence (-1.0 to 1.0)
+- arousal (0.0 to 1.0)
+Text: {content}"""
     
     data = json.dumps({
         "model": model,