sdd-apply

name	sdd-apply
description	Implement the SDD task list. Read tasks and design from memory, write code, run tests.
when_to_use	After sdd-tasks. Reads tasks, spec, and design from memory, implements each task in order.
allowed-tools	["Read","Write","Edit","Bash","Glob","Grep","mcp__invencible-memory__memory_recall","mcp__invencible-memory__memory_save"]
model	sonnet
effort	high
user-invocable	false
hub-skill-ids	["implementation","plan-execution"]

Goal

Implement all tasks from the task list, in order, verifying each one before moving to the next.

Steps

1. Load tasks: memory_recall(topic_key="sdd-<change_id>-tasks")
2. Load design: memory_recall(topic_key="sdd-<change_id>-design")
3. Load spec: memory_recall(topic_key="sdd-<change_id>-spec")
4. For each task: a. Implement the change in the specified files b. Run the project test suite — fix failures before moving on c. Verify the task's "Done when" condition is met
5. After all tasks complete, save implementation notes to memory:
   • title: SDD Apply: <change_id>
   • topic_key: sdd-<change_id>-impl
   • type: context
   • content: summary of what was changed, any deviations from the design, and test results

Rules

• Do not add features beyond the task list — stay in scope
• Run tests after every task, not just at the end
• If a task conflicts with the design, note the deviation in the impl summary — do not silently deviate
• Keep commits atomic: one task per commit