mcp-server-mas-sequential-thinking 0.2.0__tar.gz → 0.2.2__tar.gz

{mcp_server_mas_sequential_thinking-0.2.0 → mcp_server_mas_sequential_thinking-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mcp-server-mas-sequential-thinking
-Version: 0.2.0
+Version: 0.2.2
 Summary: MCP Agent Implementation for Sequential Thinking
 Author-email: Frad LEE <fradser@gmail.com>
 Requires-Python: >=3.10
@@ -146,17 +146,23 @@ The `env` section should include the API key for your chosen `LLM_PROVIDER`.
     # GROQ_TEAM_MODEL_ID="llama3-70b-8192"
     # GROQ_AGENT_MODEL_ID="llama3-8b-8192"
     # Example for DeepSeek:
-    # DEEPSEEK_TEAM_MODEL_ID="deepseek-chat"
-    # DEEPSEEK_AGENT_MODEL_ID="deepseek-coder"
+    # DEEPSEEK_TEAM_MODEL_ID="deepseek-chat"  # Note: `deepseek-reasoner` is not recommended as it doesn't support function calling
+    # DEEPSEEK_AGENT_MODEL_ID="deepseek-chat"  # Recommended for specialists
     # Example for OpenRouter:
-    # OPENROUTER_TEAM_MODEL_ID="anthropic/claude-3-haiku-20240307"
-    # OPENROUTER_AGENT_MODEL_ID="google/gemini-flash-1.5"
+    # OPENROUTER_TEAM_MODEL_ID="deepseek/deepseek-r1"
+    # OPENROUTER_AGENT_MODEL_ID="deepseek/deepseek-chat-v3-0324"
 
     # --- External Tools ---
     # Required ONLY if the Researcher agent is used and needs Exa
     EXA_API_KEY="your_exa_api_key"
     ```
 
+    **Note on Model Selection:**
+
+    *   The `TEAM_MODEL_ID` is used by the Coordinator (the `Team` object itself). This role requires strong reasoning, synthesis, and delegation capabilities. Using a more powerful model (like `deepseek-r1`, `claude-3-opus`, or `gpt-4-turbo`) is often beneficial here, even if it's slower or more expensive.
+    *   The `AGENT_MODEL_ID` is used by the specialist agents (Planner, Researcher, etc.). These agents handle more focused sub-tasks. You might choose a faster or more cost-effective model (like `deepseek-v3`, `claude-3-sonnet`, `llama3-70b`) for specialists, depending on the complexity of the tasks they typically handle and your budget/performance requirements.
+    *   The defaults provided in `main.py` (e.g., `deepseek-reasoner` for agents when using DeepSeek) are starting points. Experimentation is encouraged to find the optimal balance for your specific use case.
+
 3.  **Install Dependencies:**
 
     * **Using `uv` (Recommended):**

{mcp_server_mas_sequential_thinking-0.2.0 → mcp_server_mas_sequential_thinking-0.2.2}/README.md

@@ -127,17 +127,23 @@ The `env` section should include the API key for your chosen `LLM_PROVIDER`.
     # GROQ_TEAM_MODEL_ID="llama3-70b-8192"
     # GROQ_AGENT_MODEL_ID="llama3-8b-8192"
     # Example for DeepSeek:
-    # DEEPSEEK_TEAM_MODEL_ID="deepseek-chat"
-    # DEEPSEEK_AGENT_MODEL_ID="deepseek-coder"
+    # DEEPSEEK_TEAM_MODEL_ID="deepseek-chat"  # Note: `deepseek-reasoner` is not recommended as it doesn't support function calling
+    # DEEPSEEK_AGENT_MODEL_ID="deepseek-chat"  # Recommended for specialists
     # Example for OpenRouter:
-    # OPENROUTER_TEAM_MODEL_ID="anthropic/claude-3-haiku-20240307"
-    # OPENROUTER_AGENT_MODEL_ID="google/gemini-flash-1.5"
+    # OPENROUTER_TEAM_MODEL_ID="deepseek/deepseek-r1"
+    # OPENROUTER_AGENT_MODEL_ID="deepseek/deepseek-chat-v3-0324"
 
     # --- External Tools ---
     # Required ONLY if the Researcher agent is used and needs Exa
     EXA_API_KEY="your_exa_api_key"
     ```
 
+    **Note on Model Selection:**
+
+    *   The `TEAM_MODEL_ID` is used by the Coordinator (the `Team` object itself). This role requires strong reasoning, synthesis, and delegation capabilities. Using a more powerful model (like `deepseek-r1`, `claude-3-opus`, or `gpt-4-turbo`) is often beneficial here, even if it's slower or more expensive.
+    *   The `AGENT_MODEL_ID` is used by the specialist agents (Planner, Researcher, etc.). These agents handle more focused sub-tasks. You might choose a faster or more cost-effective model (like `deepseek-v3`, `claude-3-sonnet`, `llama3-70b`) for specialists, depending on the complexity of the tasks they typically handle and your budget/performance requirements.
+    *   The defaults provided in `main.py` (e.g., `deepseek-reasoner` for agents when using DeepSeek) are starting points. Experimentation is encouraged to find the optimal balance for your specific use case.
+
 3.  **Install Dependencies:**
 
     * **Using `uv` (Recommended):**

{mcp_server_mas_sequential_thinking-0.2.0 → mcp_server_mas_sequential_thinking-0.2.2}/README.zh-CN.md

@@ -130,17 +130,23 @@
     # GROQ_TEAM_MODEL_ID="llama3-70b-8192"
     # GROQ_AGENT_MODEL_ID="llama3-8b-8192"
     # DeepSeek 示例:
-    # DEEPSEEK_TEAM_MODEL_ID="deepseek-chat"
-    # DEEPSEEK_AGENT_MODEL_ID="deepseek-coder"
+    # DEEPSEEK_TEAM_MODEL_ID="deepseek-chat"  # 注意：不推荐使用 `deepseek-reasoner`，因为它不支持函数调用
+    # DEEPSEEK_AGENT_MODEL_ID="deepseek-chat"  # 推荐用于专家智能体
     # OpenRouter 示例:
-    # OPENROUTER_TEAM_MODEL_ID="anthropic/claude-3-haiku-20240307"
-    # OPENROUTER_AGENT_MODEL_ID="google/gemini-flash-1.5"
+    # OPENROUTER_TEAM_MODEL_ID="deepseek/deepseek-r1"
+    # OPENROUTER_AGENT_MODEL_ID="deepseek/deepseek-chat-v3-0324"
 
     # --- 外部工具 ---
     # 仅当研究员智能体被使用且需要 Exa 时才必需
     EXA_API_KEY="你的_exa_api_密钥"
     ```
 
+    **关于模型选择的说明:**
+
+    *   `TEAM_MODEL_ID` 由协调器（`Team` 对象本身）使用。该角色需要强大的推理、综合和委派能力。使用更强大的模型（如 `deepseek-r1`、`claude-3-opus` 或 `gpt-4-turbo`）通常更有益，即使它可能更慢或更昂贵。
+    *   `AGENT_MODEL_ID` 由专家智能体（规划器、研究员等）使用。这些智能体处理更集中的子任务。您可以为专家选择更快或更具成本效益的模型（如 `deepseek-v3`、`claude-3-sonnet`、`llama3-70b`），具体取决于它们通常处理的任务复杂性以及您的预算/性能要求。
+    *   `main.py` 中提供的默认值（例如，使用 DeepSeek 时的 `deepseek-chat`）是起点。鼓励进行实验，以找到适合您特定用例的最佳平衡点。
+
 3.  **安装依赖：**
 
       * **使用 `uv` (推荐):**

{mcp_server_mas_sequential_thinking-0.2.0 → mcp_server_mas_sequential_thinking-0.2.2}/main.py

@@ -313,12 +313,12 @@ def get_model_config() -> tuple[Type[Model], str, str]:
     elif provider == "groq":
         ModelClass = Groq
         team_model_id = os.environ.get("GROQ_TEAM_MODEL_ID", "deepseek-r1-distill-llama-70b")
-        agent_model_id = os.environ.get("GROQ_AGENT_MODEL_ID", "deepseek-r1-distill-llama-70b")
+        agent_model_id = os.environ.get("GROQ_AGENT_MODEL_ID", "qwen-2.5-32b")
         logger.info(f"Using Groq: Team Model='{team_model_id}', Agent Model='{agent_model_id}'")
     elif provider == "openrouter":
         ModelClass = OpenRouter
         team_model_id = os.environ.get("OPENROUTER_TEAM_MODEL_ID", "deepseek/deepseek-chat-v3-0324")
-        agent_model_id = os.environ.get("OPENROUTER_AGENT_MODEL_ID", "deepseek/deepseek-chat-v3-0324")
+        agent_model_id = os.environ.get("OPENROUTER_AGENT_MODEL_ID", "deepseek/deepseek-r1")
         logger.info(f"Using OpenRouter: Team Model='{team_model_id}', Agent Model='{agent_model_id}'")
     else:
         logger.error(f"Unsupported LLM_PROVIDER: {provider}. Defaulting to DeepSeek.")
@@ -654,7 +654,7 @@ async def sequentialthinking(thought: str, thoughtNumber: int, totalThoughts: in
              return json.dumps({
                  "error": "Critical Error: Application context not available and re-initialization failed.",
                  "status": "critical_failure"
-             }, indent=2)
+             }, indent=2, ensure_ascii=False)
              # Or raise Exception("Critical Error: Application context not available.")
 
     MIN_TOTAL_THOUGHTS = 5 # Keep a minimum suggestion
@@ -768,7 +768,7 @@ async def sequentialthinking(thought: str, thoughtNumber: int, totalThoughts: in
             "status": "success"
         }
 
-        return json.dumps(result_data, indent=2)
+        return json.dumps(result_data, indent=2, ensure_ascii=False)
 
     except ValidationError as e:
         logger.error(f"Validation Error processing tool call: {e}")
@@ -776,13 +776,13 @@ async def sequentialthinking(thought: str, thoughtNumber: int, totalThoughts: in
         return json.dumps({
             "error": f"Input validation failed: {e}",
             "status": "validation_error"
-        }, indent=2)
+        }, indent=2, ensure_ascii=False)
     except Exception as e:
         logger.exception(f"Error processing tool call") # Log full traceback
         return json.dumps({
             "error": f"An unexpected error occurred: {str(e)}",
             "status": "failed"
-        }, indent=2)
+        }, indent=2, ensure_ascii=False)
 
 # --- Main Execution ---
 

{mcp_server_mas_sequential_thinking-0.2.0 → mcp_server_mas_sequential_thinking-0.2.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "mcp-server-mas-sequential-thinking"
-version = "0.2.0"
+version = "0.2.2"
 description = "MCP Agent Implementation for Sequential Thinking"
 readme = "README.md"
 requires-python = ">=3.10"

{mcp_server_mas_sequential_thinking-0.2.0 → mcp_server_mas_sequential_thinking-0.2.2}/uv.lock

@@ -433,7 +433,7 @@ wheels = [
 
 [[package]]
 name = "mcp-server-mas-sequential-thinking"
-version = "0.2.0"
+version = "0.2.1"
 source = { editable = "." }
 dependencies = [
     { name = "agno" },