Back to skill

Security audit

Rules Tracker

Security checks for vulnerabilities and agentic risk

Overview

This is a simple rules-compliance tracking skill with disclosed reporting and local memory use, with no code execution, network access, or hidden data flow found.

Before installing, be aware that this skill may create recurring local compliance reports and memory entries when enabled. Use it when you want task-level rule tracking, and disable or narrow auto-trigger use if you do not want routine task metadata recorded.

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 (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill’s trigger guidance is broad enough that it could activate in many unrelated task contexts, causing unsolicited rule-tracking behavior at task end. In an agent setting, ambiguous activation increases the chance of prompt injection or workflow interference because an attacker can more easily induce the skill to run and influence outputs or logging without explicit user intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The auto-trigger section allows activation on vague events such as daily heartbeat, task completion keywords, or user request without defining exact matching rules or authorization boundaries. This can cause accidental or adversarial invocation, leading to unwanted persistence in memory files, leakage of task metadata, or disruption of normal agent behavior through excessive reporting.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The file presents the skill's description, usage instructions, trigger examples, and reporting template entirely in Chinese, with no indication that the user may choose another language. This creates a language-policy issue because the skill implicitly constrains interaction locale without documenting user choice or a justified region-specific requirement.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The invocation phrase '使用此 skill 追踪规则遵循情况' is generic enough that it could appear in ordinary discussion, meta-reasoning, or quoted text. That ambiguity increases the risk of accidental activation, causing the agent to switch modes, emit compliance-tracking content, or persist data when the user did not intend to call the skill.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The auto-trigger section defines broad conditions like heartbeat, task completion markers, or user requests without a narrowly scoped invocation contract. In an agent environment, this can cause the skill to activate unexpectedly during unrelated work, leading to unintended memory writes, distracting reporting behavior, or interference with higher-priority instructions.

Content

No source excerpt is available for this finding.