From 9ac531bdf023530f2921444de90744aebfe48c9b Mon Sep 17 00:00:00 2001
From: yumoqing <yumoqing@gmail.com>
Date: Thu, 7 May 2026 18:02:33 +0800
Subject: [PATCH] fix: pass user_id from .dspy to reasoning engine, remove
 hardcoded model

- reasoning_submit.dspy: pass user_id from await get_user() to hermes_reason_and_execute
- hermes_reason_and_execute: accept user_id parameter
- reason_and_execute: accept user_id parameter (removed ServerEnv.get_current_user fallback)
- _llm_call: remove hardcoded model=qwen3-max, let llm_chat use default_model from harnessed_agent_config
---
 harnessed_reasoning/core.py       | 19 +++++--------------
 wwwroot/api/reasoning_submit.dspy |  3 ++-
 2 files changed, 7 insertions(+), 15 deletions(-)

diff --git a/harnessed_reasoning/core.py b/harnessed_reasoning/core.py
index 40df9e7..49d5e59 100644
--- a/harnessed_reasoning/core.py
+++ b/harnessed_reasoning/core.py
@@ -254,18 +254,16 @@ class HermesReasoningEngine:
     async def _llm_call(self, messages: List[Dict[str, str]], config: Dict[str, Any],
                         **extra) -> Dict[str, Any]:
         """Call LLM via harnessed_agent's llm_chat."""
-        model = config.get('model_name', 'qwen3-max')
         temperature = float(config.get('temperature', 0.7))
         max_tokens = int(config.get('max_output_tokens', 4096))
 
         # Use harnessed_agent's llm_chat if available
         env = ServerEnv()
         if hasattr(env, 'llm_chat'):
-            info(f"Calling llm_chat: model={model}, temp={temperature}, max_tokens={max_tokens}")
+            info(f"Calling llm_chat: temp={temperature}, max_tokens={max_tokens}")
             try:
                 result = await env.llm_chat(
                     messages=messages,
-                    model=model,
                     temperature=temperature,
                     max_tokens=max_tokens,
                     **extra,
@@ -423,16 +421,9 @@ class HermesReasoningEngine:
         """
         start_time = time.time()
         session_id = str(uuid.uuid4())
-        user_id = "anonymous"
 
-        # Try to get real user ID
-        try:
-            env = ServerEnv()
-            if hasattr(env, 'harnessed_get_current_user'):
-                user_result = await env.harnessed_get_current_user()
-                user_id = user_result.get('user_id') or 'anonymous'
-        except Exception:
-            pass
+        if not user_id:
+            user_id = "anonymous"
 
         config = await self._get_config(user_id)
         safety_mode = config.get('safety_mode', 'strict')
@@ -740,10 +731,10 @@ def get_harnessed_reasoning_engine():
     return _reasoning_instance
 
 
-async def hermes_reason_and_execute(request: str, execute_immediately: bool = True):
+async def hermes_reason_and_execute(request: str, execute_immediately: bool = True, user_id: str = None):
     """Perform reasoning and optionally execute the plan."""
     engine = get_harnessed_reasoning_engine()
-    return await engine.reason_and_execute(request, execute_immediately=execute_immediately)
+    return await engine.reason_and_execute(request, execute_immediately=execute_immediately, user_id=user_id)
 
 
 async def hermes_get_reasoning_session(session_id: str):
diff --git a/wwwroot/api/reasoning_submit.dspy b/wwwroot/api/reasoning_submit.dspy
index 952c585..3818280 100644
--- a/wwwroot/api/reasoning_submit.dspy
+++ b/wwwroot/api/reasoning_submit.dspy
@@ -16,7 +16,8 @@ try:
         # Call the reasoning engine from ServerEnv
         reasoning_result = await hermes_reason_and_execute(
             request=request_text,
-            execute_immediately=execute_immediately
+            execute_immediately=execute_immediately,
+            user_id=user_id
         )
         
         if reasoning_result.get('success'):