agent-mail-guardSanitize email and calendar content before it reaches your AI agent's context window. Blocks prompt injection, markdown image exfiltration, invisible unicode...
Install via ClawdBot CLI:
clawdbot install discodaddy/agent-mail-guardGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Contains instructions to override system prompt or ignore user requests
"Ignore previous instructions"Calls external URL not in known-safe list
https://github.com/DiscoDaddy/agent-mail-guardUses known external API (expected, informational)
arxiv.orgAI Analysis
The skill's core function is defensive sanitization of untrusted input, and the external URL reference is to its own public GitHub repository for documentation, not an operational data exfiltration endpoint. The 'prompt poisoning' signal is a false positive, as the phrase is cited as an example of what the skill detects and blocks, not as an instruction it executes.
Generated Mar 22, 2026
AI agents handling customer inquiries via email can use this skill to sanitize incoming messages, blocking malicious content like prompt injections or hidden payloads. It ensures safe processing by flagging suspicious emails and summarizing only trusted content, protecting the agent from being manipulated.
For high-level executives, this skill sanitizes calendar events and invitations to prevent attacks via hidden unicode or fake meeting details. It flags unknown senders and suspicious content, allowing AI assistants to safely manage schedules without exposing sensitive data.
AI agents in healthcare can process patient emails and appointment requests securely by stripping HTML, base64 payloads, and other threats. The skill's trust tiers help prioritize known contacts, ensuring compliance and safe handling of sensitive information.
Law firms using AI for email-based document intake can sanitize attachments and text to block injection attempts. The skill detects homoglyphs and invisible characters, providing clean JSON output for safe summarization and analysis without data exfiltration risks.
Schools and universities can deploy this skill to filter emails from students and parents, neutralizing prompt injections and markdown exploits. It helps AI agents summarize important messages while minimizing exposure to untrusted content, enhancing security in administrative workflows.
Offer this skill as a premium add-on for existing AI agent platforms, charging a monthly fee per user or agent. It provides continuous updates for new threat patterns and integration support, generating recurring revenue from security-conscious businesses.
Sell enterprise licenses to large organizations for custom deployment and integration with their email systems like Microsoft Exchange or Google Workspace. This includes dedicated support, customization of detection patterns, and compliance features for regulated industries.
Provide the core sanitizer as open-source to build community trust, while offering paid features such as advanced analytics, cloud-based threat intelligence feeds, and priority support. This attracts users from small businesses to large enterprises, converting them to paid plans.
💬 Integration Tip
Start by testing with the provided shell scripts and adapt them to your email provider; ensure contacts.json is configured to define trusted senders for effective tier-based filtering.
Scored Jun 19, 2026
Audited Apr 17, 2026 · audit v1.0
Manage and operate ClawSec Monitor v3.0, a MITM HTTP/HTTPS proxy that logs AI agent traffic, detects exfiltration and injection threats in real time.
Scan Clawdbot and MCP skills for malware, spyware, crypto-miners, and malicious code patterns before you install them. Security audit tool that detects data exfiltration, system modification attempts, backdoors, and obfuscation techniques.
Security scanner for ClawHub skills. Vet third-party skills before installation — detect dangerous patterns, suspicious code, and risky dependencies.
Security audit and hardening for AI agents — credential hygiene, secret scanning, prompt injection defense, data leakage prevention, and privacy zones.
Scan OpenBot/Clawdbot skills for security vulnerabilities, malicious code, and suspicious patterns before installing them. Use when a user wants to audit a skill, check if a ClawHub skill is safe, scan for credential exfiltration, detect prompt injection, or review skill security. Triggers on security audit, skill safety check, malware scan, or trust verification.
OpenClaw skill discovery, security vetting & install. Searches 3000+ curated skills from ClawHub registry and awesome-openclaw-skills catalog. Scores credibility, detects prompt injection & malicious patterns, manages installations. Quick-checks GitHub for new skills.