Token导航 LogoToken导航

Agent Skill · 开发

mailgi(mailgi 开发)

mailgi 用于辅助安全审计、权限检查和凭据风险排查,适合在 OpenClaw 中需要复核安全边界、认证流程或敏感配置时使用。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。

mailgi 是开发类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为接触敏感数据;许可证为 MIT-0,第三方扫描已通过,按安装量在开发类的 21,746 个 Skill 里排第 6,043。

总安装 5,221周安装 222GitHub Stars 公开资料未说明下载 1,829核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.2

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:mailgi(mailgi 开发)
来源仓库:https://github.com/oyagev/mailgi
安装命令:
openclaw skills install mailgi
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 mailgi

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

注册和管理电子邮件代理,以通过 Mailgi 的 API 发送、接收、组织和阅读电子邮件,并具有可选的基于区块链的计费和 DID 身份验证。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install mailgi;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

mailgi 怎么安装?

在终端执行 openclaw skills install mailgi。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

mailgi 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

mailgi 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

mailgi 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

mailgi 用的人多吗?

公开数据是累计安装 5,221 次,近一周安装 222 次,下载 1,829 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

mailgi 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理开发类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
mailgi

mailgi 开发 · 本页

5,221—接触敏感数据MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

mailgi — SKILL FILE

This file teaches you how to use the mailgi email API. You are an AI agent. Read this file, then you can send and receive email.

Base URL: https://api.mailgi.xyz Auth: Authorization: Bearer <apiKey> on all authenticated requests.


1. Get an email address

Register once. No password, no OAuth.

POST /v1/agents/register
Content-Type: application/json

{ "label": "my-agent" }

Response:

{
  "agentId": "clxxx...",
  "emailAddress": "buzzing-falcon@mailgi.xyz",
  "aliasAddress": "x7k3mwf2qr5b@mailgi.xyz",
  "apiKey": "amb_...",
  "apiKeyId": "clyyy..."
}

Store apiKey immediately. It is shown exactly once.

emailAddress is your friendly address. Use it for sending and tell others to send to it. aliasAddress is a deterministic alias — both receive mail to the same inbox.


2. Check your profile

GET /v1/agents/me
Authorization: Bearer <apiKey>

3. Read your inbox

GET /v1/mail
Authorization: Bearer <apiKey>

Optional query params:

  • mailboxId — filter to a specific folder
  • limit — max results (default 20, max 100)
  • position — pagination offset (default 0)
  • sort — asc or desc (default desc)

Response: { messages: [...], total: N, position: N }

Each message has: id, subject, from, to, receivedAt, preview, seen.

Get full body of a message:

GET /v1/mail/<id>
Authorization: Bearer <apiKey>

Response includes htmlBody and/or textBody. If body is a string, use it directly. If it is an array of JMAP parts, look up bodyValues[part.partId].value for the text.


4. Send email

POST /v1/mail/send
Authorization: Bearer <apiKey>
Content-Type: application/json

{
  "to": ["someone@example.com"],
  "subject": "Hello from my agent",
  "textBody": "Hi there."
}

Optional fields: cc, bcc, htmlBody, replyTo. to, cc, bcc accept a single string or an array of strings.

Response: { "messageId": "..." }

Sending is free. Rate limit: 100 external emails per day per API key.


5. Manage mailboxes (folders)

List folders:

GET /v1/mailboxes
Authorization: Bearer <apiKey>

Each mailbox has id, name, role (inbox/sent/trash/drafts/etc), totalEmails, unreadEmails.

Create a folder:

POST /v1/mailboxes
Authorization: Bearer <apiKey>
Content-Type: application/json

{ "name": "Projects", "parentId": "<optional parent id>" }

Move a message to a folder:

PATCH /v1/mail/<id>/move
Authorization: Bearer <apiKey>
Content-Type: application/json

{ "mailboxId": "<folder id>" }

Mark as read:

PATCH /v1/mail/<id>/flags
Authorization: Bearer <apiKey>
Content-Type: application/json

{ "seen": true }

Delete a message (moves to Trash):

DELETE /v1/mail/<id>
Authorization: Bearer <apiKey>

6. API keys

You can create additional API keys (e.g. one per task):

POST /v1/apikeys
Authorization: Bearer <apiKey>
Content-Type: application/json

{ "label": "task-runner", "expiresAt": "2026-12-31T00:00:00Z" }

Response includes apiKey (raw, shown once) and id.

List keys: GET /v1/apikeys Revoke a key: DELETE /v1/apikeys/<keyId>


7. DID-based auth (optional)

If you registered with a did:key: DID, you can authenticate without an API key:

  1. Request a challenge:
POST /v1/auth/challenge
Content-Type: application/json

{ "did": "did:key:z6Mk..." }
  1. Sign the returned nonce with your Ed25519 private key (base64url), then verify:
POST /v1/auth/verify
Content-Type: application/json

{ "did": "did:key:z6Mk...", "nonce": "...", "signature": "<base64url Ed25519 sig>" }

Response: { "token": "...", "expiresIn": 3600 } — use as Authorization: Bearer <token>.


8. Error responses

All errors follow:

{ "error": { "code": "ERROR_CODE", "message": "Human-readable description" } }

Common codes:

  • 401 — missing or invalid API key
  • 404 — message or mailbox not found
  • 409 — conflict (e.g. mailbox name already exists)
  • 429 — rate limit exceeded

9. Health

GET /health        — liveness (always 200 if server is up)
GET /health/ready  — readiness (checks DB + mail server)

Quick start (copy-paste)

# 1. Register
RESP=$(curl -s -X POST https://api.mailgi.xyz/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"label":"my-agent"}')
EMAIL=$(echo $RESP | jq -r .emailAddress)
KEY=$(echo $RESP | jq -r .apiKey)

# 2. Send a message
curl -s -X POST https://api.mailgi.xyz/v1/mail/send \
  -H "Authorization: Bearer $KEY" \
  -H "Content-Type: application/json" \
  -d "{\"to\":[\"someone@example.com\"],\"subject\":\"Hi\",\"textBody\":\"Hello from $EMAIL\"}"

# 3. Read inbox
curl -s https://api.mailgi.xyz/v1/mail \
  -H "Authorization: Bearer $KEY"

10. TypeScript / Node.js SDK

Install:

npm install @mailgi/mailgi
import { AgentMailboxClient } from '@mailgi/mailgi';

// Construct from a stored API key
const client = AgentMailboxClient.withApiKey(
  'https://api.mailgi.xyz',
  process.env.MAILGI_API_KEY!,
);

// Register a new agent (first time only — save the returned apiKey)
const reg = await client.agents.register({ label: 'my-agent' });
// reg.emailAddress => 'buzzing-falcon@mailgi.xyz'
// reg.apiKey       => 'amb_...'  (shown once — store it)
client.apiKey = reg.apiKey;

// Send email
const { messageId } = await client.mail.send({
  to: ['someone@example.com'],
  subject: 'Hello',
  textBody: 'Hi from my agent.',
});

// Read inbox
const { messages } = await client.mail.list({ limit: 20, sort: 'desc' });
const email = await client.mail.get(messages[0].id);
console.log(email.subject, email.textBody);

// Mark as read
await client.mail.setFlags(email.id, { seen: true });

All SDK methods map 1-to-1 to the REST endpoints above. Errors extend AgentMailboxError with statusCode and code:

import { NotFoundError, UnauthorizedError } from '@mailgi/mailgi';

try {
  await client.mail.get('bad-id');
} catch (err) {
  if (err instanceof NotFoundError) console.error('Not found');
  if (err instanceof UnauthorizedError) console.error('Bad API key');
}

11. CLI

Install globally:

npm install -g @mailgi/mailgi

All commands require --agent <email-or-username>.

# Register a new agent (saves API key to ~/.mailgi/config.json)
mailgi register --label my-agent --agent me@mailgi.xyz

# Save an existing agent by API key
mailgi login --agent buzzing-falcon@mailgi.xyz --apikey amb_...

# List saved agents
mailgi agents

# Show agent profile (live from API)
mailgi me --agent buzzing-falcon

# Read inbox
mailgi inbox --agent buzzing-falcon
mailgi inbox --agent buzzing-falcon --limit 50

# Read a message (auto-marks as seen)
mailgi read --agent buzzing-falcon <message-id>

# Send email
mailgi send --agent buzzing-falcon --to alice@example.com --subject "Hi" --body "Hello"
mailgi send --agent buzzing-falcon --to alice@example.com --subject "Hi" --body-file ./message.txt

# Delete a message
mailgi delete --agent buzzing-falcon <message-id>

# Mailboxes
mailgi mailboxes --agent buzzing-falcon
mailgi mailboxes create "Projects" --agent buzzing-falcon
mailgi mailboxes delete <id> --agent buzzing-falcon

# API keys
mailgi keys --agent buzzing-falcon
mailgi keys create --label task-key --agent buzzing-falcon
mailgi keys revoke <key-id> --agent buzzing-falcon

# Config
mailgi config show
mailgi config set-url https://api.mailgi.xyz

# Remove saved agent
mailgi logout --agent buzzing-falcon --yes

# Raw JSON output (any command)
mailgi inbox --agent buzzing-falcon --json

Full interactive docs: https://api.mailgi.xyz/docs Machine-readable spec: https://api.mailgi.xyz/openapi.json


Support

Questions or issues? Email objective-crocodile@mailgi.xyz — yes, it's a real mailgi inbox.

平台分布

OpenClaw

96.03%
按下载量换算1,756

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills