Back to skill

Security audit

test

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only email intelligence skill whose sensitive email handling is expected for its purpose and not paired with hidden execution or install behavior.

Install only if you intend to work with authorized email data. Treat mailbox content, attachments, and indexed search results as sensitive, and ensure any real implementation adds access control, tenant scoping, audit logging, retention rules, and redaction before connecting it to live email systems.

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 (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

This markdown file describes fetching raw email messages, parsing headers, bodies, and attachments from an email source, which can expose sensitive personal or enterprise data. The surrounding skill description does not provide any warning or disclosure about privacy impact, mailbox access, or handling sensitive content.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill exposes email_ask and email_search interfaces that can retrieve and return message content from indexed threads without any described authorization checks, query scoping, sensitivity filtering, or user-facing warning. In an email-intelligence context, these tools make it easy for downstream agents to surface confidential mailbox data, attachments, or internal communications to unauthorized prompts, users, or tenants if integrated incautiously.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation guidance is broad enough that merely referencing the agent by name or specialty could invoke it unintentionally. In multi-agent or prompt-routed environments, this can cause incorrect agent selection, context leakage, or execution of specialized behavior when the user did not explicitly request it.

Content

No source excerpt is available for this finding.