Back to skill

Security audit

Messages

Security checks for vulnerabilities and agentic risk

Overview

The skill is openly about message management, but it asks to aggregate and profile sensitive communications across channels with limited scoping and retention controls.

Review this skill carefully before installing. It may be useful, but it needs access to highly sensitive email, chat, and messaging history, including sent messages and relationship context. Only connect accounts and channels you are comfortable having analyzed, and look for controls to exclude sensitive contacts, limit history, approve actions before they happen, and delete any stored profiles or cross-channel mappings.

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 (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill describes broad autonomous handling of incoming messages and daily briefings without clearly constraining what data sources, message types, or user approvals are required before access and triage occur. In a messaging skill, ambiguous activation and scope can lead to over-collection of sensitive communications and unintended processing of private or regulated content.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill explicitly promises a complete picture of incoming messages across connected channels, but it does not warn users that private content from email, chat, and messaging platforms will be aggregated and analyzed centrally. This creates significant privacy and confidentiality risk because sensitive personal, business, legal, or regulated communications could be ingested without users appreciating the breadth of exposure.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill states that it reads existing sent messages to model the user's writing style, but it does not disclose the privacy implications of profiling prior communications. Sent mail often contains highly sensitive information, and using it for voice profiling can expose confidential content, infer relationship patterns, and create secondary-use risks beyond the user's likely expectations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill describes maintaining a unified view of people across multiple channels and surfacing full conversation history regardless of source, but it does not warn about identity correlation and cross-context aggregation. This is especially dangerous because combining email, chat, social, and messaging records can reveal relationship graphs, sensitive associations, and contextual information users may expect to remain separated by platform.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.