{"skill":{"slug":"email-security","displayName":"Email Security","summary":"Protect AI agents from email-based attacks including prompt injection, sender spoofing, malicious attachments, and social engineering. Use when processing emails, reading email content, executing email-based commands, or any interaction with email data. Provides sender verification, content sanitization, and threat detection for Gmail, AgentMail, Proton Mail, and any IMAP/SMTP email system.","tags":{"agent-safety":"1.0.0","attachments":"1.0.0","dkim":"1.0.0","dmarc":"1.0.0","email":"1.0.0","eml":"1.0.0","gmail":"1.0.0","imap":"1.0.0","latest":"1.0.0","phishing":"1.0.0","prompt-injection":"1.0.0","sanitization":"1.0.0","security":"1.0.0","spf":"1.0.0","spoofing":"1.0.0","threat-detection":"1.0.0"},"stats":{"comments":0,"downloads":1164,"installsAllTime":0,"installsCurrent":0,"stars":2,"versions":1},"createdAt":1770576621482,"updatedAt":1777525047306},"latestVersion":{"version":"1.0.0","createdAt":1770576621482,"changelog":"# 🛡️ Email Security Skill (v1.0.0)\n\n**Protect your AI agent from email-based attacks including prompt injection, sender spoofing, malicious attachments, and social engineering.**\n\nThis unified skill provides a comprehensive security layer for any agent handling email, supporting Gmail, AgentMail, Proton, and generic IMAP/SMTP providers.\n\n## 🚀 Key Features\n\n*   **🚫 Prompt Injection Defense**: Automatically detects and blocks 15+ advanced attack patterns (e.g., \"Ignore previous instructions\", role hijacking, hidden text).\n*   **👤 Sender Verification**: Validates sender identity using SPF/DKIM/DMARC checks and an customizable Owner/Admin/Trusted whitelist.\n*   **🧹 Smart Content Sanitization**: Strips dangerous HTML, removes tracking pixels, and extracts only the newest message content (ignoring safe quoted replies).\n*   **📎 Safe Attachments**: Enforces strict file type allowances (PDF, TXT, CSV, Images) and blocks executables/scripts.\n*   **🔒 Provider Agnostic**: Works seamlessly with Gmail, AgentMail, and any standard email service.\n\n## 🛡️ Security Capabilities\n\n- **Role-Based Access Control**:\n    - **Owner**: Full system control.\n    - **Admin**: Operational commands.\n    - **Trusted**: Standard interactions.\n    - **Unknown**: Blocked by default.\n- **Threat Intelligence**: Built-in detection for social engineering cues (urgency, financial requests).\n- **Audit Logging**: Comprehensive logs for all blocked and flagged interactions.\n\n## 📦 What's Included\n\n- **Scripts**: `verify_sender.py`, `sanitize_content.py`, `parse_email.py`\n- **Documentation**: Full setup guides for Gmail & AgentMail.\n- **Configuration**: JSON templates for custom security policies.\n\n## 🔧 Quick Start\n\n```python\n# Verify sender before processing\nauth_result = verify_sender(email=\"user@example.com\")\n\n# Sanitize content to remove threats\nclean_body = sanitize_content(raw_email_body)\n\n# Parse allowed attachments safely\nfiles = parse_email(raw_message_data)\n```\n\n## 📋 Version 1.0.0 Release Notes\n\n- Added role-based authorization model (Owner/Admin/Trusted/Unknown).\n- Implemented heuristic spoofing detection (From/Reply-To mismatches).\n- Added multi-stage content sanitization pipeline.\n- Enforced strict attachment security policies.\n- Packaged with full provider support documentation.","license":null},"metadata":null,"owner":{"handle":"ivaavimusic","userId":"s179r9z3n07jxsdszy7pdwqans83tjq8","displayName":"Ivaavi.eth","image":"https://avatars.githubusercontent.com/u/99599836?v=4"},"moderation":{"isSuspicious":true,"isMalwareBlocked":false,"verdict":"suspicious","reasonCodes":["suspicious.prompt_injection_instructions"],"summary":"Detected: suspicious.prompt_injection_instructions","engineVersion":"v2.4.5","updatedAt":1777525047306}}