Back to skill

Security audit

Email Agent Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a planning and review aid for agent-managed email workflows and does not contain code, hidden execution, persistence, or credential handling.

Install with ordinary caution for a workflow-planning skill. Users should still verify that agents follow the stated approval gates before live sends, contact imports, DNS changes, suppression edits, or production automation changes.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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 (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger description says to use the skill when an AI agent needs to 'plan, review, implement, audit, or improve email work' and then lists broad concepts like 'tool handoffs' and 'delegation plans.' Although some examples are provided, the activation scope remains expansive and lacks exclusion conditions or negative examples, increasing the chance of unintended invocation for ordinary email-related coordination tasks.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The manifest uses the activation phrase "Use $emailagentskill to help me plan, audit, or improve an email workflow," which includes broad intents like "help me plan" and "improve" without clear scope boundaries or exclusion conditions. In a manifest file, this can create ambiguity about when the skill should activate for ordinary email-related requests versus when it should not.

Content

No source excerpt is available for this finding.