Gives your agent the ability to create structured, AI-optimized implementation plans for new features, refactoring, or infrastructure upgrades, ensuring deterministic and autonomous execution.
When you need a clear, phase-by-phase roadmap for a complex technical task
When you want to ensure an AI agent can execute your implementation without ambiguity
When you need to document the rationale, dependencies, and testing strategy for a new feature
Define the purpose of your plan, such as a feature update or a system refactor
List the specific requirements, constraints, and patterns that must be followed
Allow the agent to break the work into discrete, atomic implementation phases
Review and save the generated plan to your project's /plan/ directory
你
Create an implementation plan for upgrading our database from v14 to v16.
Agent
I'll generate upgrade-database-v16-1.md in your /plan/ directory. I'll structure it into three phases: Environment Prep, Data Migration, and Validation. Each phase will include specific TASK codes with exact shell commands and file references, ensuring the entire upgrade can be executed systematically with zero ambiguity.
赋予您的代理使用机器可读结构，针对新需求、重构或升级更新现有实现计划的能力。
赋予你的智能体获取实现计划并自动生成一系列 GitHub Issue 的能力，项目中的每个阶段对应一个 Issue。
Gives your agent the ability to act as a veteran software engineer, creating detailed feature implementation plans with system diagrams, database schemas, and API designs.
Gives your agent the ability to generate comprehensive, standardized technical documentation for object-oriented components, complete with architecture diagrams and API references.
Gives your agent the ability to define clear, unambiguous, and AI-ready specifications for schemas, tools, architecture, and more, ensuring a machine-readable project foundation.
