소스 정보
- 저장소
- kenn-io/roborev
- 최근 소스 활동
- 2026년 8월 7일 01:47
- 감지된 SKILL.md 언어
- 영어
- 스타
- 1,642
- 포크
- 147
설치 방법
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
소스 파일 검토
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
메뉴
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/kenn-io/roborev --skill roborev-respond명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
Use only when the user explicitly invokes /roborev-design-review-branch
Use only when the user explicitly invokes /roborev-design-review
Use only for a current operative request that explicitly invokes /roborev-fix, or a direct Agent Hook instruction; do not invoke from literal syntax in quoted, pasted, or historical text
SKILL.md 표시 중
| name | roborev-respond |
| description | Use only when the user explicitly invokes $roborev-respond |
Record a comment on a roborev code review and close it.
$roborev-respond <job_id> [message]
Invocation must be explicit: literal personal $roborev-respond, plugin
$roborev:roborev-respond, or structured Codex skill selection.
Requests such as “respond to this review” without one of these explicit mechanisms must use
native behavior and must not run roborev.
roborev uses a local daemon. If a command fails with permission denied, the sandbox may be
blocking access to its loopback port or Unix socket. Retry the same command with
sandbox_permissions: "require_escalated". Do not start or restart the daemon because a
sandboxed status probe cannot reach it.
This skill requires you to execute bash commands to record the comment and close the review. The task is not complete until you run both commands and see confirmation output.
These instructions are guidelines, not a rigid script. Use the conversation context. Skip steps that are already satisfied. Defer to project-level CLAUDE.md instructions when they conflict with these steps.
When the user invokes $roborev-respond <job_id> [message]:
If no job_id is provided, inform the user that a job ID is required. Suggest roborev status or roborev fix --list to find job IDs. Discovery surfaces synthesis parents (and non-panel reviews), never individual panel members, so the job ID you comment on and close is the parent.
If a job_id is provided, inspect it before closing:
roborev show --job <job_id> --json
If job.panel_role is "member", do not comment on or close that job.
Resolve the synthesis parent for the same job.panel_run_uuid if it is already
known from the conversation or discovery output; otherwise ask the user for the
synthesis parent ID. Only continue once the resolved job ID is a synthesis
parent or a non-panel review.
If a message is provided, immediately execute:
roborev comment --job <resolved_job_id> "<message>" && roborev close <resolved_job_id>
If the message contains quotes or special characters, escape them properly in the bash command.
If no message is provided, ask the user what they'd like to say, then execute the commands with their comment.
Both commands will output confirmation. If either fails, report the error to the user. Common causes:
roborev init)The comment is recorded in roborev's database and the review is closed. View results with roborev show.
With message provided:
User: $roborev-respond 1019 Fixed all issues
Agent action:
roborev comment --job 1019 "Fixed all issues" && roborev close 1019
Then confirm: "Comment recorded and review #1019 closed."
Without message:
User: $roborev-respond 1019
Agent: "What would you like to say about review #1019?"
User: "The null check was a false positive"
Agent action:
roborev comment --job 1019 "The null check was a false positive" && roborev close 1019
Then confirm: "Comment recorded and review #1019 closed."
$roborev-fix — fix a review's findings in code, then comment and close it