Feature request for product/service
Cursor IDE
Describe the request
When generating plans in cursor it would be excellent if we could have much more control over what happens at each stage of a plan.
Being able to activate skills or sub-agents at different stages would be excellent to tailor for different aspects of development.
For example:
- Have backend python agent run development steps (with attached skills)
- Then activate testing agents (again with specific backend testing skills).
- Then have overall ai-review who can evaluate quality, test results and whether code quality is good.
Depending on the plan or the task we can activate specialized agents with skills rather than having one model try and bundle through a plan.