Sutra

Instructions

Analyzes a task to recommend the most efficient tool (The Router).

Args:
    task_description: The user's prompt or task.

Implementation Reference

• src/context_engineering_mcp/server.py:180-244 (handler)

  The primary handler function for the 'analyze_task_complexity' tool. It is decorated with @mcp.tool() for automatic registration with the FastMCP server. The function validates input using TaskComplexityInput schema, analyzes keywords in the task description, and returns a recommendation for strategy, complexity, tool, and reasoning.

  @mcp.tool()
  def analyze_task_complexity(task_description: str) -> dict:
      """
      Analyzes a task to recommend the most efficient tool (The Router).
  
      Args:
          task_description: The user's prompt or task.
      """
      try:
          model = TaskComplexityInput(task_description=task_description)
      except ValidationError as e:
          return {"error": str(e)}
  
      task = model.task_description.lower()
  
      # Strategy: Constructor Mode (Build/Design)
      if any(
          w in task
          for w in [
              "build",
              "create",
              "design",
              "architect",
              "system",
              "bot",
              "assistant",
          ]
      ):
          return {
              "strategy": "constructor",
              "complexity": "Variable",
              "recommended_tool": "design_context_architecture",
              "reasoning": "User wants to build a system/agent. Use the Architect to design a blueprint.",
          }
  
      # Strategy: YOLO Mode (Direct Solve)
      if any(w in task for w in ["project", "repo", "codebase", "architecture"]):
          return {
              "strategy": "yolo",
              "complexity": "Medium",
              "recommended_tool": "project.explore",
              "reasoning": "Task involves project-level understanding.",
          }
      elif any(w in task for w in ["test", "tdd", "verify"]):
          return {
              "strategy": "yolo",
              "complexity": "High",
              "recommended_tool": "workflow.test_driven",
              "reasoning": "Task involves testing or verification workflows.",
          }
      elif any(w in task for w in ["analyze", "reason", "think", "solve", "complex"]):
          return {
              "strategy": "yolo",
              "complexity": "High",
              "recommended_tool": "reasoning.systematic",
              "reasoning": "Task requires structured reasoning.",
          }
      else:
          return {
              "strategy": "yolo",
              "complexity": "Low",
              "recommended_tool": "Standard Molecule",
              "reasoning": "Task appears simple. Use a basic prompt or few-shot molecule.",
          }

• src/context_engineering_mcp/server.py:36-40 (schema)

  Pydantic BaseModel schema used for input validation in the analyze_task_complexity tool handler.

  class TaskComplexityInput(BaseModel):
      task_description: str = Field(
          ..., min_length=5, description="The user's prompt or task."
      )