generateContextPrompt
Creates detailed text prompts summarizing a plan's current state for AI decision-making, including goals, ongoing tasks, and actionable next steps within MCPlanManager.
Instructions
生成一个详细的文本提示，总结计划的当前状态。 这个提示可以作为上下文提供给AI模型，以帮助其决定下一步行动。 内容包括：总体目标、当前任务、可执行任务列表等。
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Implementation Reference
- src/mcplanmanager/app.py:195-205 (handler)MCP tool handler and registration for 'generateContextPrompt'. This function creates a DependencyPromptGenerator instance using the global plan_manager and calls its generate_context_prompt method to produce the context prompt.@mcp.tool() def generateContextPrompt() -> str: """ 生成一个详细的文本提示，总结计划的当前状态。 这个提示可以作为上下文提供给AI模型，以帮助其决定下一步行动。 内容包括：总体目标、当前任务、可执行任务列表等。 """ from .dependency_tools import DependencyPromptGenerator generator = DependencyPromptGenerator(plan_manager) prompt = generator.generate_context_prompt() return prompt
- Core helper function in DependencyPromptGenerator class that implements the logic to generate the context prompt by collecting plan data, status, tasks, dependencies, and suggestions.def generate_context_prompt(self) -> str: """生成上下文感知的提示词""" plan_status = self.pm.getPlanStatus() if not plan_status["success"]: return "Error: Could not get plan status" dump_result = self.pm.dumpPlan() if not dump_result.get("success"): return "Error: Could not dump plan data" plan_data = dump_result["data"] goal = plan_data["meta"]["goal"] tasks = plan_data["tasks"] state = plan_data["state"] prompt_parts = [ "# 任务执行上下文", f"## 总体目标\n{goal}", "", "## 当前状态" ] # 当前任务信息 current_task_response = self.pm.getCurrentTask() if current_task_response["success"]: task = current_task_response["data"] prompt_parts.extend([ f"- 当前执行任务: [{task['id']}] {task['name']}", f"- 任务状态: {task['status']}", f"- 执行理由: {task['reasoning']}" ]) else: prompt_parts.append("- 当前没有活动任务") # 可执行任务 executable = self.pm.getExecutableTaskList() if executable["success"] and len(executable["data"]) > 0: prompt_parts.append("\n## 可执行任务") for task in executable["data"]: prompt_parts.append(f"- [{task['id']}] {task['name']}") # 任务依赖关系 prompt_parts.extend([ "", "## 任务依赖关系", self._generate_dependency_text(tasks) ]) # 执行建议 prompt_parts.extend([ "", "## 执行建议", self._generate_execution_suggestions(tasks, state) ]) return "\n".join(prompt_parts)