Engineering AI Skills

Explore user intent, requirements and design before implementation

Create detailed implementation plans from specs or requirements

Create isolated git worktrees with smart directory selection

Execute implementation plans in a separate session with review checkpoints

Supervise and delegate to agents in controlled batches, track progress

Execute implementation plans with independent tasks in current session

Run multiple independent tasks in parallel without shared state

Write tests before implementation code

Debug methodically with evidence before proposing fixes

Verify work meets requirements before merging

Process feedback with technical rigor, not blind implementation

Run verification commands and confirm output before success claims

Complete development work with structured options for merge, PR, or cleanup

Establish how to find and use skills in any conversation

Create new skills, edit existing skills, verify skills work