ソース情報
- リポジトリ
- htlin222/dotfiles
- ソースの最終更新活動
- 2026年7月26日 12:13
- 検出された SKILL.md の言語
- 英語
- スター
- 79
- フォーク
- 4
インストール方法
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
ソースファイルを確認
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
メニュー
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/htlin222/dotfiles --skill mailコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか?SkillsMP が現在取得できるファイルをダウンロードできます。
依 ticket/issue 產出初版實作 — 解析需求、從最新 develop 切出符合命名規範的分支、寫出實作、跑既有測試與 lint、conventional commit,然後交棒給 /simplify。Use when the user types /develop, or asks to start implementing a ticket, issue, or feature request end-to-end from requirement to first commit.
Put a website behind a Cloudflare Access (Zero Trust) login gate, or remove one, entirely from the CLI — no dashboard GUI. Use when the user wants to password/email-protect a hostname, gate a Cloudflare Pages or Workers site, restrict a site to specific emails, set up Zero Trust Access, or asks about "cf-gate". Manages Access applications and allow-email policies via the Cloudflare API using a token in the skill's .env. Note: wrangler does NOT manage Access — this uses the Cloudflare REST API directly.
Curate a GitHub repository's wiki (the separate repo.wiki.git) into a coherent, tightly written set of pages: Home, Introduction (project + features), Roadmap, Gotchas/Lessons, Tech Debt, and an Architecture page taught through the book *Head First Software Architecture* — with mermaid diagrams, in the repo's own language and style, then commit and push. Use when the user wants to create, update, curate, or document a GitHub repo's wiki; write or refresh wiki pages; add an architecture / design page to a wiki; enable a wiki; or asks for "/wiki-git". Handles both first-time wikis and updates to existing ones, and can fan out across many repos.
SOC 職業分類に基づく
SKILL.md を表示中
| name | |
| description | Read Mail.app inboxes and create Reminders for actionable items. Use for email triage. |
Reads macOS Mail.app, identifies actionable emails, and creates reminders in Reminders.app "Inbox" list.
/mail — process past 24 hours (default)/mail 7 days — process past 7 days/mail 14 days — process past 14 daysIssue ALL of these tool calls in a single parallel message:
Bash — Activate apps + fetch emails (combined to avoid extra round-trip):
osascript -e 'tell application "Mail" to activate' && osascript -e 'tell application "Reminders" to activate' && sleep 1 && osascript ~/.claude/skills/mail/scripts/fetch-mail.applescript <days>
<days> is 1. If user provides args like "7 days", "48h", or "14 days", extract the number.id, account, subject, from, date, read, preview.Read — Load dedup database from /tmp/mail-already-processed.json:
/tmp/mail-already-processed.json with: {"processed_ids":[],"last_run":""}Bash — Fetch existing reminders for dedup:
~/.claude/skills/mail/scripts/reminders-cli fetch
After Batch A completes, filter out emails whose id is already in processed_ids. Only analyze NEW emails.
For each new email, determine if it requires action. Classify into:
Actionable (create reminder):
Not actionable (skip):
For each actionable email, extract:
Issue ALL extract-urls.sh calls in a single parallel message — one Bash tool call per actionable email:
~/.claude/skills/mail/scripts/extract-urls.sh "<message-id>"
Returns one URL per line, filtered to remove tracking pixels, images, and social media links.
How to use extracted URLs:
Performance note: Only run this for actionable emails (typically 1-5 per batch). Each call takes ~2-3s but they run in parallel, so total wall time ≈ one call.
Before adding, check if a similar reminder already exists (from Batch A's reminders fetch) by comparing:
Skip any reminder that would be a duplicate.
Use the batch-add command to create all reminders in a single process invocation:
printf '%s' '<JSON_ARRAY>' | ~/.claude/skills/mail/scripts/reminders-cli batch-add
Where <JSON_ARRAY> is a JSON array of objects:
[
{
"name": "Task title",
"body": "Details and context",
"due": "YYYY-MM-DD HH:MM",
"priority": "1"
},
{
"name": "Another task",
"body": "More details",
"due": "YYYY-MM-DD HH:MM",
"priority": "5"
}
]
Priority values: "1" (high/urgent), "5" (medium/normal), "9" (low/FYI).
Returns OK:<count> on success.
Shell escaping: The JSON must be valid. Use printf '%s' to pipe JSON to stdin. Avoid single quotes inside the JSON — use escaped double quotes for string values.
Fallback: If batch-add fails, fall back to individual reminders-cli add calls — issue them ALL in a single parallel message:
~/.claude/skills/mail/scripts/reminders-cli add "<name>" "<body>" "<due_date>" "<priority>"
After processing, update /tmp/mail-already-processed.json:
processed_idslast_run timestampUse the Write tool (not bash heredoc/redirect) to write the updated JSON:
{
"processed_ids": ["id1", "id2", ...],
"last_run": "2026-03-10T00:00:00"
}
Important: Do NOT use cat > file << EOF — the check-file-exists hook will block the > redirect. Always use the Write tool for this file.
Present results to user in a table:
## Mail Summary (past <N> days)
**Processed**: X new emails across Y accounts
**Skipped**: Z already-processed emails
### Actionable Items Added to Reminders:
| # | Due | Reminder | Priority | Source | Link |
|---|-----|----------|----------|--------|------|
| 1 | 3/9 12:00 | Task name | HIGH | sender@email.com | [action link](url) |
### Non-Actionable (skipped):
- Newsletter from X
- Notification from Y
If no actionable URL was found for an item, show "—" in the Link column.
/tmp/mail-already-processed.json persists across reboots only if /tmp is not cleaned. This is acceptable — reprocessing is harmless since we also check existing reminders.