Back to skill

Security audit

My Skill

Security checks for vulnerabilities and agentic risk

Overview

This is a minimal testing skill that describes local data-processing helpers but does not include executable code or hidden behavior.

Before installing, understand that the skill is very general and should only be used on files you intentionally provide or identify. Ask for confirmation before any file writes, deletes, or broad workspace automation if future versions add real tools.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The usage text says the agent can invoke the skill whenever it 'needs to process custom text, manage local files, or handle specific workflow automation,' which is extremely broad and underspecified. Ambiguous activation criteria can cause an agent to use the skill in unintended situations, especially around local file access, increasing the chance of unauthorized file operations or unsafe automation without clear user consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill advertises local file management capability but does not warn users or the agent about the sensitivity of local file access, the scope of files that may be touched, or the need for confirmation before modifying data. In agent environments, missing safety boundaries around local file behavior can lead to unintended reads, overwrites, or workflow actions on user files.

Content

No source excerpt is available for this finding.