subagent-driven-development

To autonomously execute a development plan, this Skill breaks it down into individual tasks and dispatches a dedicated subagent for each, ensuring rapid and high-quality development cycles.

To prevent file conflicts during parallel code execution, this Skill uses worktree isolation, ensuring filesystem isolation for each subagent and enabling safe concurrent development without conflicts.

Automated code review for subagent-driven development uses a two-stage process, rigorously checking both spec compliance and code quality after each task is completed by a subagent.

Yes, this autonomous task execution approach integrates directly with deployment and user testing workflows, managing the transition from completed code review to downstream release stages.

Yes, you can execute independent development tasks in parallel, as the Skill dispatches fresh subagents for each task in a plan, allowing for both parallel and sequential processing.

To start automating feature plan execution, you need a detailed feature plan with multiple independent development tasks, such as a markdown file located at a specified path.