Back to skill

Security audit

技能优化器

Security checks for vulnerabilities and agentic risk

Overview

This skill is not malicious, but it can persistently change other skills and its activation and approval boundaries are too broad for that level of authority.

Install only in an environment where you are comfortable with a skill that can modify other installed skills. Before use, require the agent to name the target files, show a diff, and get explicit approval before writing; narrowing the triggers would reduce accidental activation risk.

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 (6)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The natural-language content and trigger phrases are entirely specified in Chinese, and the file does not indicate that users may choose another language or that Chinese is a required locale for a justified regional use case. This can constitute a language/locale policy issue when the skill behavior is effectively restricted to one language without opt-in.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes broad everyday phrases such as '修改技能', '改进技能', and '技能有问题', which can plausibly appear in normal conversation and unintentionally activate this skill. Because this skill modifies other skills, accidental invocation could lead to unauthorized or unintended changes to agent behavior, making the context more dangerous than a read-only helper skill.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger set is broad enough to activate on ordinary conversation about improving skills, retrospectives, or discussing past mistakes, which can cause the agent to enter a file-modifying workflow without a clearly scoped user intent. In this skill’s context, that is more dangerous because the skill is explicitly designed to edit SKILL.md and related records, so accidental invocation can lead to unintended persistent changes to agent behavior.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill describes optimization behavior but does not clearly warn that it will modify persistent skill files, version history, and pitfalls records. In this context, the omission is risky because users may think they are only discussing improvements, while the skill can perform lasting configuration changes that alter future agent behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger examples are broad everyday phrases like “修改技能” and “技能有问题”, which can overlap with normal conversation and cause the optimizer skill to activate unintentionally. In this skill’s context, accidental activation is meaningful because the skill is authorized to modify other skills and documentation, so a false trigger could lead to unintended changes or workflow disruption.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

SQP-3 适用于所有文件类型。该文件通篇以中文给出规范和示例,且未说明这是可选语言或面向特定中文场景的限定,可能构成未经过用户选择的语言/locale 默认约束。

Content

No source excerpt is available for this finding.