ClawHub
Back to skill

Security audit

Check AgentMailer Email

Security checks for vulnerabilities and agentic risk

Overview

This skill is a read-only AgentMailer email helper whose sensitive email access is disclosed and bounded by user-directed triage instructions.

Install this only if you want the agent to read AgentMailer inbox content through the AgentMailer MCP service. Because email, drafts, and attachments can contain sensitive information, users should invoke it for specific inbox review tasks and avoid asking it to fetch attachments unless needed.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The tool description grants access to drafts and attachments, which is broader than the skill’s stated purpose of inbox/thread review and summarization. This scope expansion can expose sensitive unsent content and file data to the agent, increasing the chance of over-collection or misuse if the skill is invoked for routine email triage.

External Transmission

Medium
Category
Data Exfiltration
Content
value: "agentmailer"
      description: "Read and search AgentMailer inboxes, messages, threads, drafts, and attachments"
      transport: "streamable_http"
      url: "https://api.agentmailer.ai/mcp"
policy:
  allow_implicit_invocation: true
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Static analysis

No suspicious patterns detected.