subagent-driven-development

Read the plan file once, extract all tasks into a todo list, then dispatch a fresh implementer subagent per task with the full task text in context. Follow each implementation with a spec compliance review and a code quality review before marking the task complete.

The first stage checks spec compliance against the original task requirements, catching under-building or scope creep. The second stage reviews code quality including conventions, error handling, test coverage, and security. Spec review must pass before quality review begins.

A fresh subagent per task prevents context pollution from accumulated state, giving each task clean focused context without confusion from prior tasks' code or reasoning. The trade-off is more subagent invocations, but issues are caught earlier.

Each task should represent 2-5 minutes of focused work, such as creating a single model or endpoint. Broad tasks like implementing an entire authentication system should be split into discrete units like model creation, password hashing, and token generation.

The implementer subagent or a new fix subagent addresses the specific issues, then the reviewer re-reviews the result. This loop repeats until approval, and the workflow never proceeds to the next task while either review has open issues.

Avoid it when you have no implementation plan, when tasks are tightly coupled and touch the same files, or when you would skip the review loops. Dispatching parallel implementers on shared files and accepting close-enough spec compliance are explicit anti-patterns.