Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/shimo4228/claude-harness/git-workflownpx skills add shimo4228/claude-harness --skill git-workflowgit clone --depth 1 https://github.com/shimo4228/claude-harnessWrote 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/shimo4228/claude-harness/git-workflow)<a href="https://agentmods.dev/skills/shimo4228/claude-harness/git-workflow"><img src="https://agentmods.dev/badge/skills/shimo4228/claude-harness/git-workflow.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00108 | $0.01036 |
| Opus 5 | $0.00054 | $0.00518 |
| Sonnet 5 | $0.00022 | $0.00207 |
| Haiku 4.5 | $0.00011 | $0.00104 |
Grade B, and why
git-workflow scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
許可設定は `Bash(git:*)`（`~/.claude/settings.json` の permissions.allow）。照合エンジンは What it actually says
git-workflow — この環境での git コマンド実行作法
許可設定は Bash(git:*)（~/.claude/settings.json の permissions.allow）。照合エンジンは
複合コマンドを分解して各セグメントを独立照合する（区切り: && || ; | & 改行）。
全セグメントが git なら自動許可されるが、以下の罠は allowlist で回避できない。
規則
- git 同士の連結（
git add … && git commit …等）は可 — 各セグメントがBash(git:*)に載るため auto-approve される（2026-08-13 実測 + 公式 permissions doc）。 非 git コマンドを 1 つでも挟むとそのセグメントで照合が外れる - cd を git と混ぜない —
git -C <dir>を使う。cd + git の組み合わせは、双方が 個別に許可済みでもハードコードの特例としてプロンプトする（移動先ディレクトリの git hooks 実行リスクのため。公式 permissions doc 明記）。Bash(cd:*)を足しても回避不能 - コミットメッセージは
-m "…"の単純形（複数-m可）または-F <file>。 バッククォート・$( )・heredoc は injection 検出で必ず承認要求になる （公式 security doc: 「Suspicious bash commands require manual approval even if previously allowlisted」。なお$( )を含む catastrophic removal は 2.1.208 以降--dangerously-skip-permissionsでも昇格する — 一般の$( )はこの特例の対象外） git pushは sandbox の network / credential 制約で失敗するためdangerouslyDisableSandbox: trueを付けて実行する （認証はgh auth setup-git済み — memory: github-auth-git-gh-disconnect-2026-06）- commit 前の secret scan は PreToolUse hook が自動実行する（rules/common/security.md）。 手動で scan を連結する必要はない。連結 commit でも hook は全 git ターゲットを走査する （ADR-0038）が、迷ったら commit だけ単発にするのが安全側
規則 2 と 3 は hook が機械的に強制する。hooks/validate-bash.sh（settings.json の
PreToolUse に配線）が cd ... && git 形と git commit + $( 形を block し、書き直しを
指示する（テスト: tests/validate-bash-cd-git.bats / tests/validate-bash-heredoc-commit.bats）。
覚えていなくても踏み外せないが、block されてから直すと 1 往復損する。
適用外
- git 以外のコマンド同士の連結は本 skill の対象外（同じ分解照合の原理は働くので、 全セグメントが許可済み prefix に載っていれば通る）
失効条件
本 skill は Claude Code の permission 実装（v2.1.x、as-of 2026-08-13）に依存する。 cd + git 特例の緩和・injection 検出の変更・sandbox 内 push の解禁があれば該当規則を再監査する。 規則がすべて product / allowlist 側に吸収されたら skill ごと退役する。
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.
- 4d ago First seen · 48 lines · 108 tokens per session scan B 3d00ad43c8eb
git-workflow is a skill published in the GitHub repository shimo4228/claude-harness (2 stars, last pushed 5d ago), licensed MIT. It adds 108 tokens to every session and 1,036 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
commit
Atomic git commit with conventional message. Use when the user says "commit", "save my changes", "commit this", or wants to create a git commit. Stages specific files, writes a conventional commit message with body explaining non-obvious decisions. Never uses git add -A.
git-workflow
Git best practices, branching strategies, commit conventions, and PR workflows. Use when reviewing git history, writing commits, setting up branching strategy, or improving git practices. Triggers on "git best practices", "commit message", "branching strategy", or "PR workflow".
git-commit-convention
写 git 提交信息、整理提交历史时使用——Conventional Commits 的格式、类型选择和拆分原则。.
zod
Zod v4 best practices, patterns, and API guidance for schema validation, parsing, error handling, and type inference in TypeScript applications. Covers safeParse, object composition, refinements, transforms, codecs, branded types, v3→v4 migration, and testing schemas with Jest or Vitest. Baseline: zod ^4.3.0. Triggers…
akigitcommit
Analyze the working tree and commit changes in clean logical groups. Triages a long half-finished tree first (finished vs mid-edit vs abandoned vs accidental) before grouping. Auto-detects CHANGELOG to switch between domain-grouped mode (3–5 commits by object/feature) and type-grouped mode (feat/fix/refactor). Stages…
conventional-commit
Draft a Conventional Commit message when the user requests a commit or message. Formats only; does not authorize Git operations.