Token导航 LogoToken导航TokenDH.com
研究检索只读clawhub未标认证来源可访问clear审计通过

email-inbox电子邮件收件箱

Agent Skill

email-inbox 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

3,515

周安装

145

GitHub Stars

公开资料未说明

下载量

1,148
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:email-inbox(电子邮件收件箱)
来源仓库:https://github.com/fangweifen2025-dotcom/email-inbox
安装命令:
openclaw skills install email-inbox
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install email-inbox

简介

对收件箱邮件进行总结、分类和起草操作建议,辅助用户高效管理邮件。

  • 适合希望 AI 代理主动检查未读消息、识别重要沟通的需求。
  • 可自动标记关键邮件、生成回复草稿或提出处理建议。email-inbox 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 需授权访问邮箱权限,建议限制扫描频率以保护隐私。
  • 使用前请核实是否支持目标邮箱类型及维护更新状态。

SKILL.md

name
email-inbox-triage
description
summarize, triage, and draft actions for incoming email. use when the user wants chatgpt to check an inbox, identify important unread messages, group emails by urgency or topic, extract action items, prepare reply drafts, or help process new mail with gmail or another connected mail source.

Email Inbox Triage

Overview

Use this skill to process incoming email in a consistent way: check the inbox, identify what matters, summarize the important parts, and prepare the next action.

Core workflow

Follow this sequence unless the user asks for only one specific step.

  1. Identify the mailbox source.

- Prefer a connected mail tool such as Gmail when available. - If no connector is available, ask the user to provide exported emails, pasted email text, or screenshots.

  1. Collect the target messages.

- If the user says "new mail" or "recent email", focus on unread or recent inbox messages first. - If the user names a sender, company, subject, or date range, filter to those messages. - If the user asks for "important" email, infer importance from sender, urgency, deadlines, invoices, approvals, customer requests, and direct asks.

  1. Triage the messages.

- Classify each message into one of these buckets: - urgent - needs reply - needs review - informational - low priority - Detect explicit deadlines, meeting times, payment requests, approvals, and follow-up asks.

  1. Produce a useful output.

Default to this structure unless the user asks for another format:

### Inbox summary - total messages reviewed - unread count when available - 1-3 sentence overall summary

### Priority items For each important email include: - sender - subject - why it matters - requested action - deadline or time sensitivity

### Suggested next steps - quick bullets for what to do next

  1. Draft replies when requested.

- Keep drafts short, clear, and ready to send. - Match the sender's tone when appropriate. - If information is missing, leave clear placeholders instead of inventing facts.

Decision rules

  • Prefer concise summaries over long paraphrases.
  • Surface risk early: deadlines, money, access, legal/compliance, customer escalations, or executive requests.
  • Do not mark newsletters or automated notifications as urgent unless the content clearly requires action.
  • When multiple related emails appear, collapse them into one thread-level summary.
  • If the user asks what to do first, recommend the smallest high-impact action.

Output templates

Template: inbox triage

Use this default format:

Inbox summary

[short overall summary]

Priority items

  1. [subject] — [sender]

- Why it matters: [reason] - Action needed: [action] - Timing: [deadline or "no explicit deadline"]

Other messages

  • [short grouped summary]

Recommended next steps

  • [step 1]
  • [step 2]

Template: reply draft

Use this when the user asks for a reply:

Subject: [keep or revise as needed]

Hi [name],

[clear response]

[requested next step or confirmation]

Best, [sign-off]

Connector guidance

  • When a mail connector is available, search the inbox before answering from memory.
  • Read the relevant messages before drafting a reply.
  • If the user asks for counts like unread inbox totals, prefer label or folder counts when the mail tool supports them.
  • If the request refers to attachments, inspect attachment metadata and read the attachment when possible.

What not to do

  • Do not invent email content, sender intent, or missing context.
  • Do not send email automatically unless the user explicitly asks for sending.
  • Do not expose raw internal ids or system metadata to the user.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

91.99%
按下载量换算1,056

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills