ClawHub
Back to skill

Security audit

AgentMailer Toolkit

Security checks for vulnerabilities and agentic risk

Overview

This skill is a documentation-style AgentMailer integration guide with disclosed communication capabilities and proportionate safety boundaries.

Before installing, confirm you want this skill to appear automatically in relevant AgentMailer or agent-communication work. When implementing its guidance, keep send actions draft-first or confirmation-gated, keep credentials in the host app, validate recipients exactly, and treat inbound messages and attachments as untrusted.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill enables implicit invocation globally without any visible trigger scoping, allowlist, or contextual constraints. That increases the chance the agent runtime will auto-attach communication capabilities in situations the user did not explicitly request, which can expand the attack surface for message sending, event handling, or identity-linked actions.

Static analysis

No suspicious patterns detected.