commit-changes
About
The commit-changes skill stages, commits, and amends Git changes using conventional commit messages. It enables selective staging, descriptive message writing via HEREDOC, and commit history verification. Use it when saving a logical unit of work to version control or creating/amending a commit with a proper conventional message.
Quick Install
Claude Code
Recommendednpx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/commit-changesCopy and paste this command in Claude Code to install this skill
Documentation
提交變更
選擇性暫存文件、撰寫清晰的提交訊息,並驗證提交歷史。
適用時機
- 將一個邏輯工作單元保存至版本控制
- 以描述性、符合慣例的訊息建立提交
- 修訂最近一次提交(訊息或內容)
- 提交前審查將納入提交的變更
輸入
- 必要:一個或多個待提交的變更文件
- 選擇性:提交訊息(若未提供則代為擬定)
- 選擇性:是否修訂前次提交
- 選擇性:共同作者標註
步驟
步驟一:審查當前變更
檢查工作樹狀態並察看差異:
# See which files are modified, staged, or untracked
git status
# See unstaged changes
git diff
# See staged changes
git diff --staged
預期: 清晰呈現所有已修改、已暫存與未追蹤的文件。
失敗時: 若 git status 失敗,驗證當前是否位於 git 倉庫內(git rev-parse --is-inside-work-tree)。
步驟二:選擇性暫存文件
以具名方式暫存特定文件,避免使用 git add . 或 git add -A,以免意外納入敏感文件或無關變更:
# Stage specific files by name
git add src/feature.R tests/test-feature.R
# Stage all changes in a specific directory
git add src/
# Stage parts of a file interactively (not supported in non-interactive contexts)
# git add -p filename
提交前審查已暫存內容:
git diff --staged
預期: 僅有預期的文件與變更被暫存。無 .env、憑證或大型二進制文件。
失敗時: 以 git reset HEAD <file> 取消暫存誤加的文件。若敏感資料已暫存,提交前立即取消暫存。
步驟三:撰寫提交訊息
採用慣例式提交格式。務必透過 HEREDOC 傳入訊息以確保格式正確:
git commit -m "$(cat <<'EOF'
feat: add weighted mean calculation
Implements weighted_mean() with support for NA handling and
zero-weight filtering. Includes input validation for mismatched
vector lengths.
Co-Authored-By: Claude Opus 4.6 <[email protected]>
EOF
)"
慣例式提交類型:
| 類型 | 使用時機 |
|---|---|
feat | 新功能 |
fix | 錯誤修復 |
docs | 僅文件 |
test | 新增或更新測試 |
refactor | 既不修復亦不新增功能之代碼變更 |
chore | 構建、CI、依賴更新 |
style | 格式、空白(無邏輯變更) |
預期: 提交已建立,訊息具描述性,說明為何而非僅何事。
失敗時: 若 pre-commit 鉤子失敗,修復問題,以 git add 重新暫存,並建立新提交(勿用 --amend,因失敗的提交從未建立)。
步驟四:修訂最後一次提交(選擇性)
僅在提交尚未推送至共享遠端時修訂:
# Amend message only
git commit --amend -m "$(cat <<'EOF'
fix: correct weighted mean edge case for empty vectors
EOF
)"
# Amend with additional staged changes
git add forgotten-file.R
git commit --amend --no-edit
預期: 前次提交已就地更新。git log -1 顯示修訂後的內容。
失敗時: 若提交已推送,勿修訂。改建新提交。對共享分支強制推送已修訂提交會造成歷史分歧。
步驟五:驗證提交
# View the last commit
git log -1 --stat
# View recent commit history
git log --oneline -5
# Verify the commit content
git show HEAD
預期: 提交以正確的訊息、作者與文件變更出現於歷史中。
失敗時: 若提交包含錯誤文件,用 git reset --soft HEAD~1 撤銷提交同時保留暫存變更,再重新正確提交。
驗證
- 僅預期文件納入提交
- 無敏感資料(權杖、密碼、
.env文件)被提交 - 提交訊息遵循慣例式提交格式
- 訊息主體說明變更之因由
-
git log顯示提交具正確元資料 - Pre-commit 鉤子(如有)已通過
常見陷阱
- 一次提交過多:每次提交應代表一個邏輯變更。將無關變更拆分為獨立提交。
- 盲目使用
git add .:務必先審查git status。優先以具名方式暫存特定文件。 - 修訂已推送提交:切勿修訂已推送至共享分支的提交。此舉改寫歷史,為協作者帶來困擾。
- 模糊的提交訊息:「修復錯誤」或「更新」毫無訊息量。描述變更了什麼及為何。
- 內容修訂時遺漏
--no-edit:將遺漏文件加入最後一次提交時,用--no-edit保留既有訊息。 - 鉤子失敗後使用
--amend:當 pre-commit 鉤子失敗,提交從未建立。使用--amend將修改前次提交。修復鉤子問題後應建立新提交。
相關技能
manage-git-branches— 提交前的分支工作流程create-pull-request— 提交後的下一步resolve-git-conflicts— 處理合併或 rebase 時的衝突configure-git-repository— 倉庫設置與慣例
GitHub Repository
Related Skills
content-collections
MetaThis skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.
polymarket
MetaThis skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.
creating-opencode-plugins
MetaThis skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
