subagent-driven-development
Executes plan tasks through fresh subagents per task, keeping the orchestrating session's context clean and each task's context focused.
Anatomy
| Parent repo | obra/superpowers |
|---|---|
| Path | skills/subagent-driven-development |
| Trigger conditions | "Use when executing implementation plans with independent tasks in the current session" |
| Instruction format | yaml-frontmatter-md |
| Required tools | subagents |
| Reusability | universal |
| Complexity | high |
| Ships scripts / examples / tests / boundaries | no / yes / no / yes |
Failure modes
Per-task context assembly is overhead; tasks with heavy shared context pay it repeatedly.
Verdict
Applies the fresh-context principle from review to execution itself. The orchestrator-worker split it teaches is where the whole ecosystem is heading.
Related skills
Blocks the agent from claiming work is complete, fixed, or passing until it has run verification commands and confirmed output. Evidence…
Forces root-cause investigation before any fix: reproduce, hypothesize one cause at a time, test the hypothesis, then patch. Bans…
Frequently asked questions
What does the subagent-driven-development skill do?
Executes plan tasks through fresh subagents per task, keeping the orchestrating session's context clean and each task's context focused.
When does subagent-driven-development trigger?
Its stated trigger: "Use when executing implementation plans with independent tasks in the current session". Trigger precision is a scored reliability dimension.