Executing Plans

A written plan is like a recipe. You need to follow each step carefully. This skill helps an AI agent execute a plan one step at a time. It starts by loading the plan and reviewing it for any problems. If something is unclear, the agent stops and asks for help.

Each task gets checked off after it is done. The agent runs verifications to make sure everything works. If a task fails repeatedly, the agent stops. Never guess. Always ask for clarification. This keeps the work safe and correct.

When all tasks are done, the agent uses another skill to finish the development branch. This whole process prevents mistakes and follows the plan exactly. It is ideal for anyone who wants reliable step-by-step implementation.