Borrowing it
Nothing to install: this file belongs to byh3071-cpu/vhk. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/byh3071-cpu/vhk/main/.agents/skills/auto-merge/SKILL.mdgit clone --depth 1 https://github.com/byh3071-cpu/vhkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/byh3071-cpu/vhk/auto-merge)<a href="https://agentmods.dev/skills/byh3071-cpu/vhk/auto-merge"><img src="https://agentmods.dev/badge/skills/byh3071-cpu/vhk/auto-merge/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/byh3071-cpu/vhk/auto-merge"><img src="https://agentmods.dev/badge/skills/byh3071-cpu/vhk/auto-merge.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00025 | $0.01533 |
| Opus 5 | $0.00013 | $0.00766 |
| Sonnet 5 | $0.00005 | $0.00307 |
| Haiku 4.5 | $0.00003 | $0.00153 |
Grade A, and why
auto-merge scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
auto-merge — 제한된 자동 머지
가동: 전용 Codex 세션에서
$auto-merge를 명시적으로 호출한다. 호출 1회는 열린 PR을 한 번만 훑고 종료한다. 헌법 관계: 머지까지만 자동. npm publish는 사람 2FA(가드 #119), main 직접 push 금지 그대로.
0. 하드 스톱 (매 주기 첫 단계, 예외 없음)
- 저장소 루트에서
.vhk/HARD_STOP존재 여부를 읽기 전용으로 확인한다. - 존재 → 즉시 전체 중단, 사유 보고. 우회·삭제 금지.
- 원래 checkout은 상태 확인 외에 이동하거나 수정하지 않는다. G4에서만 정확한 PR head를 detached 임시 worktree로 분리한다.
1. 대상 수집
gh pr list --state open --base main --label auto-merge --limit 1000 --json number,title,headRefName,headRefOid,baseRefName,additions,deletions
gh repo view --json nameWithOwner
auto-merge라벨 없는 PR은 절대 건드리지 않는다 (조회도 라벨 필터로만).- 전체 대상 목록을 수집한 다음 주기당 최대 3개 제한을 적용한다. 각 PR의
headRefOid는 이번 주기의 검토·머지 기준으로 유지한다. - 0건이면 "대상 없음" 한 줄 보고 후 주기 종료.
2. PR별 게이트 (순서대로, 하나라도 탈락 시 스킵)
| 게이트 | 판정 | 탈락 시 |
|---|---|---|
| G1. CI | gh pr checks <n> 전부 pass. pending 있으면 "대기" — 실패 아님, 다음 주기 재검 |
스킵 |
| G2. diff 상한 | additions+deletions ≤ 500줄. 초과 = 사람 검토 필요 | 스킵 + 보고에 "사람 호출" 표기 |
| G3. CodeRabbit 해소 | 아래 GraphQL로 미해결 리뷰 스레드 0개 확인 | 스킵 |
| G4. 적대적 최종 리뷰 | 아래 §3 | 스킵 + 사유 보고 |
G3 쿼리 — nameWithOwner를 owner/name으로 나누고 GraphQL 변수로 전달한다. 저장소 좌표를 파일에 하드코딩하지 않는다. 현재 셸의 따옴표 규칙만 사용하고 서로 다른 셸 문법을 섞지 않는다.
query($owner:String!, $name:String!, $number:Int!, $endCursor:String) {
repository(owner:$owner, name:$name) {
pullRequest(number:$number) {
reviewThreads(first:100, after:$endCursor) {
nodes { isResolved }
pageInfo { hasNextPage endCursor }
}
}
}
}
hasNextPage가 false가 될 때까지 endCursor를 다음 요청에 넘긴다. 어느 페이지든 isResolved: false가 하나라도 있으면 탈락한다. 조회 실패나 페이지 누락은 통과가 아니라 스킵이다.
3. 적대적 최종 리뷰 (G4)
원래 checkout에서 base와 PR head를 갱신한 뒤, PR head를 detached 임시 worktree로 분리해 독립 리뷰를 실행한다. <temporary-path>는 이번 실행만을 위한 새 경로여야 한다.
git update-ref -d refs/vhk-auto-merge/pr-<n>
git fetch origin main "pull/<n>/head:refs/vhk-auto-merge/pr-<n>"
git rev-parse refs/vhk-auto-merge/pr-<n>
git rev-parse 결과가 수집 시 저장한 <headRefOid>와 같은 경우에만 아래 단계로 진행한다.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 99 lines · 25 tokens per session scan A 22c4afa6c627
auto-merge is a skill published in the GitHub repository byh3071-cpu/vhk (12 stars, last pushed 3d ago), licensed MIT. It adds 25 tokens to every session and 1,533 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
git-release
Draft release notes from merged PRs, propose a semver bump, and emit a copy-pasteable gh release create command. Use when preparing a tagged GitHub release.
ship-pr
Run the project checks and open a pull request once they pass.
github-operations
Unified GitHub interaction covering artifact types, comment conventions, CLI commands, and issue lifecycle. Used when reading from or writing to GitHub Issues, managing the project board, or posting structured comments.
project-memory
GitHub-native persistence for project learnings, decisions, and patterns using issue labels and structured comments. Used when recording what worked, what failed, or architectural decisions that should persist across sessions.
maxsim-batch
Orchestrates parallel agent execution using worktree isolation following Anthropic's batch pattern. Used when multiple independent tasks can be executed simultaneously.
commit-conventions
Enforces conventional commit format with atomic changes and co-author attribution. Used when committing code changes to maintain consistent git history.