Back to skill

Security audit

Dmaic Process Improvement

Security checks for vulnerabilities and agentic risk

Overview

This is a visible DMAIC process-improvement guide with no executable code, but users should avoid putting sensitive business data into its attachments or generated reports.

Install only if you want a Chinese DMAIC/Six Sigma guided workflow. Before uploading files or saving/exporting reports, remove customer, employee, financial, regulated, proprietary, or unnecessary confidential details, and be aware that broad terms like quality analysis or Six Sigma may invoke the skill.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger phrases include very generic terms such as “质量分析” and “六西格玛”, which can plausibly appear in normal conversation and unintentionally activate the skill. Because this skill drives multi-step guided interactions and document generation, accidental invocation could confuse users, derail unrelated sessions, or cause unintended collection of business/process information.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

该描述将“流程改进”“质量分析”列为触发词,这些短语在企业讨论、咨询或普通问答中都非常常见,缺少足够的域内约束,容易与用户的一般性表达重叠。文档也未提供明确的排除条件或负例来说明何种情况下不应激活该技能。

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly invites users to upload files and states that the AI can read text content, but it does not warn users against including confidential, regulated, or personal data. In a DMAIC/quality context, attachments may contain production data, customer complaints, employee information, or proprietary process documents, so the lack of privacy handling guidance increases the risk of unintended sensitive-data exposure.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

“我想做流程改进”属于宽泛自然语言表达,既可表示普通咨询,也可能被解释为显式调用技能。这里仅给出示例而未说明是否仅限这些短语、是否需要明确命令式触发,导致激活范围不清。

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill repeatedly generates, stores, and exports detailed project documentation without warning that these materials may include sensitive operational metrics, financial estimates, customer feedback, and internal process weaknesses. In this business-improvement context, automatic persistence and export increase the chance of oversharing or retaining sensitive data beyond what the user expects.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

文件中的技能名称说明、触发词、示例回复和全部交互模板均固定为中文,未表明可根据用户偏好切换语言。若组织要求避免未经选择地强制特定语言,这构成语言/locale 选择不足的问题。

Content

No source excerpt is available for this finding.