← 返回 Skills 市场
cameron48

Email Engine

作者 Cameron48 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
109
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install mcf-email-engine
功能描述
Send emails, bulk campaigns, AI-powered classification, auto-reply generation, and template management.
使用说明 (SKILL.md)

Email Engine

Send transactional and outreach emails, run bulk campaigns, classify inbound messages with AI, generate contextual replies, and manage reusable templates. Designed for AI agents that need to communicate with humans over email as part of autonomous workflows.

Authentication

All endpoints require x402 payment (USDC on Base L2). Send a request without payment to receive pricing info in the 402 response.

Endpoints

Send Email

  • Method: POST
  • Path: /api/email/send
  • Price: $0.02 per call
  • Description: Send a single email. Supports HTML and plain text, attachments, and reply-to threading.

Request:

{
  "to": "[email protected]",
  "from": "[email protected]",
  "subject": "Quick question about your scheduling, Mike",
  "body_html": "\x3Cp>Hi Mike,\x3C/p>\x3Cp>I noticed Comfort Zone HVAC is crushing it on reviews...\x3C/p>",
  "body_text": "Hi Mike, I noticed Comfort Zone HVAC is crushing it on reviews...",
  "reply_to": "[email protected]",
  "track_opens": true,
  "track_clicks": true
}

Response:

{
  "id": "msg_4k9xr2",
  "status": "sent",
  "to": "[email protected]",
  "sent_at": "2026-04-01T14:40:00Z"
}

Send Bulk

  • Method: POST
  • Path: /api/email/send-bulk
  • Price: $0.01 per recipient
  • Description: Send a campaign to multiple recipients. Supports personalization variables and staggered delivery.

Request:

{
  "template_id": "tmpl_cold_outreach_v2",
  "recipients": [
    {"email": "[email protected]", "variables": {"first_name": "Mike", "company": "Comfort Zone HVAC"}},
    {"email": "[email protected]", "variables": {"first_name": "Sarah", "company": "North Texas Air"}}
  ],
  "from": "[email protected]",
  "stagger_minutes": 5,
  "track_opens": true
}

Response:

{
  "campaign_id": "camp_7n3mp1",
  "total_recipients": 2,
  "status": "queued",
  "estimated_completion": "2026-04-01T14:50:00Z"
}

AI Classify

  • Method: POST
  • Path: /api/email/ai/classify
  • Price: $0.05 per call
  • Description: Classify an inbound email using AI. Returns intent, sentiment, urgency, and suggested action.

Request:

{
  "subject": "Re: Quick question about your scheduling",
  "body": "Hey Cameron, this sounds interesting. We've been struggling with scheduling for a while. Can you tell me more about pricing?",
  "sender": "[email protected]"
}

Response:

{
  "classification": {
    "intent": "interested",
    "sentiment": "positive",
    "urgency": "medium",
    "category": "sales-reply",
    "suggested_action": "send-pricing",
    "confidence": 0.94
  }
}

AI Reply

  • Method: POST
  • Path: /api/email/ai/reply
  • Price: $0.10 per call
  • Description: Generate a contextual reply to an email thread using AI. Takes conversation history and CRM context.

Request:

{
  "thread": [
    {"from": "[email protected]", "body": "Quick question about your scheduling..."},
    {"from": "[email protected]", "body": "This sounds interesting. Can you tell me more about pricing?"}
  ],
  "context": {
    "lead_stage": "qualified",
    "solution": "ai-scheduling-bundle",
    "monthly_price": "$497"
  },
  "tone": "casual-professional",
  "goal": "book-a-call"
}

Response:

{
  "reply": {
    "subject": "Re: Quick question about your scheduling",
    "body": "Hey Mike,\
\
Glad it caught your eye. The AI scheduling package runs $497/mo and typically pays for itself within the first month...",
    "call_to_action": "calendar-link"
  }
}

Manage Templates

  • Method: POST
  • Path: /api/email/templates
  • Price: $0.02 per call
  • Description: Create or update reusable email templates with personalization variables.

Request:

{
  "name": "Cold Outreach v2",
  "subject": "Quick question about your {{pain_point}}, {{first_name}}",
  "body_html": "\x3Cp>Hi {{first_name}},\x3C/p>\x3Cp>I noticed {{company}} is doing great work in {{industry}}...\x3C/p>",
  "variables": ["first_name", "company", "industry", "pain_point"]
}

Response:

{
  "id": "tmpl_cold_outreach_v2",
  "name": "Cold Outreach v2",
  "variables": ["first_name", "company", "industry", "pain_point"],
  "created_at": "2026-04-01T14:45:00Z"
}

Use Cases

  • An outreach agent sends personalized cold emails to leads from the pipeline
  • A support agent classifies inbound emails and routes them to the right workflow
  • An autonomous sales agent generates and sends follow-up replies based on prospect responses
  • A marketing agent runs drip campaigns across segmented lead lists
  • An agent orchestrator monitors email threads and escalates urgent messages

Pricing

Endpoint Price Description
/api/email/send $0.02 Send a single email
/api/email/send-bulk $0.01/recipient Bulk campaign send
/api/email/ai/classify $0.05 AI email classification
/api/email/ai/reply $0.10 AI reply generation
/api/email/templates $0.02 Create/update templates
安全使用建议
This skill is internally consistent, but before installing you should: 1) Verify the gateway (https://gateway.mcfagentic.com) — look for documentation, a privacy policy, and reputation information — because all email content, recipient lists, and CRM context will be sent there. 2) Confirm billing mechanics for the x402 payment flow so you understand charges and can limit spending. 3) Test with safe addresses and small batches to ensure the gateway enforces 'from' address verification and rate limits (to avoid spoofing or accidental spam). 4) Restrict autonomous use in your agent: require explicit approval for bulk sends or sending to real prospects. 5) Check compliance with applicable anti-spam/privacy rules (CAN-SPAM, GDPR) before running campaigns. If you need stronger assurance, ask the publisher for a homepage, documentation, or an audit of their sending practices and provenance of the gateway service.
功能分析
Type: OpenClaw Skill Name: mcf-email-engine Version: 1.0.0 The skill bundle defines a standard API interface for an email service provider (MCF Agentic) hosted at gateway.mcfagentic.com. It includes endpoints for sending emails, bulk campaigns, and AI-driven classification/replies, all of which align with the stated purpose. No malicious instructions, data exfiltration attempts, or prompt injection vulnerabilities were identified in SKILL.md or _meta.json.
能力评估
Purpose & Capability
Name/description (send emails, campaigns, AI classification/reply, templates) align with the endpoints documented in SKILL.md. No unrelated binaries, env vars, or config paths are requested; the gateway handles sending and billing, which explains the lack of local credentials.
Instruction Scope
SKILL.md instructs callers to POST email content, recipients, thread history, and CRM context to gateway.mcfagentic.com endpoints — this is consistent with the skill purpose. Note: those instructions will transmit potentially sensitive email content and recipient lists to an external service; the file does not instruct the agent to read local files or extra environment variables beyond what's in the request payload.
Install Mechanism
Instruction-only skill with no install spec or code files — lowest install risk. Nothing is downloaded or written to disk by the skill itself.
Credentials
No environment variables or credentials are requested by the skill. This is coherent because authentication/billing is handled by an external x402 payment flow described in the SKILL.md. However, sending emails typically requires a sending identity; here the gateway accepts arbitrary 'from' addresses in requests which could enable spoofing if the gateway doesn't enforce verification.
Persistence & Privilege
The skill is not always-enabled and uses normal autonomous invocation behavior. It does not request persistent system privileges or modify other skills' configs.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install mcf-email-engine
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /mcf-email-engine 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
- Initial release of Email Engine skill. - Enables sending transactional and outreach emails, including bulk campaigns. - AI-powered inbound email classification and automatic contextual reply generation. - Supports creation and management of email templates with personalization. - All endpoints require x402 payment (USDC on Base L2).
元数据
Slug mcf-email-engine
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Email Engine 是什么?

Send emails, bulk campaigns, AI-powered classification, auto-reply generation, and template management. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 109 次。

如何安装 Email Engine?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install mcf-email-engine」即可一键安装,无需额外配置。

Email Engine 是免费的吗?

是的,Email Engine 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Email Engine 支持哪些平台?

Email Engine 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Email Engine?

由 Cameron48(@cameron48)开发并维护,当前版本 v1.0.0。

💬 留言讨论