subagent-check-runner

Builds the list of automated checks a coding task requires, dispatches a single subagent to run tests, lint, type checks and builds outside the main conversation, and drives a fix-and-rerun loop until the list is green, while holding the task open for human acceptance of anything automation cannot judge. Use when a change has just been implemented and the task's plan or acceptance criteria name checks that must pass before it counts as finished. Not for writing new tests, reviewing code by reading it, continuous monitoring, or shipping the verified change.

Version

v1.0.0

Size

14 KB

Updated

2026-10-11

Your client verifies the signature before reading. This explicit manual read is available on every plan. Free automatically loads five chosen skills; paid plans automatically load relevant skills across the library.