원클릭으로
mts-one
Use when a declared frontend, backend, or integration workstream has one READY task for parent-led sequential TDD.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Use when a declared frontend, backend, or integration workstream has one READY task for parent-led sequential TDD.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | mts-one |
| description | Use when a declared frontend, backend, or integration workstream has one READY task for parent-led sequential TDD. |
Perform one workstream task through parent-led sequential TDD. This is the default execution mode and does not require a native writing subagent.
status: READY, matching lane/workstream, and a matching
current branch/worktree. The parent Agent manually creates or enters that
worktree; this Skill never runs Git setup.IN_PROGRESS in the sidecar and record the parent host
and identity in the track execution metadata.For one focused behavior at a time:
Update the task Evidence, Review, and Result sections with commands, exit
codes, platform, and material results. Record NOT RUN with a reason where
appropriate. Move the task through IN_REVIEW to DONE, or BLOCKED with a
specific owner and unblock condition. The parent Agent is the only maintainer
of task records and commits.
Use when an active frontend, backend, or integration workstream must pause and record a sidecar HANDOFF.md.
Use when a paused frontend, backend, or integration workstream must resume from its sidecar TRACK.md and HANDOFF.md records.
Use when the user explicitly requests a complete Electron workflow rather than a small edit or one-stage request.
Use when the current host has dated WRITE_SUBAGENT: VERIFIED evidence and the user requests parent-supervised native-subagent TDD for one workstream task.
Use when an approved Electron contract must split one feature into frontend, backend, and integration workstreams with TDD task records.
Use when an Electron failure or confirmed review finding can be reproduced and needs root-cause analysis, a minimal fix, and regression verification. Do not use for speculative cleanup, broad refactoring, initial review, or masking failures by weakening safeguards.