Back to skill

Security audit

胡田 OPC导师 三会治理系统

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a business-governance advisory framework, but it should be reviewed because it includes broad automatic activation plus government-influence and personnel-evaluation guidance without enough safeguards.

Install only if you want a Chinese-language business-governance advisory framework and are prepared to supervise its outputs. Do not treat its government-relations, HR, legal, finance, or crisis recommendations as final decisions; require human compliance/legal review, avoid providing unnecessary personal or confidential data, and keep sensitive public-sector or personnel actions outside automatic workflows.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (28)

Tp2

High
Category
MCP Tool Poisoning
Confidence
85% confidence
Finding

Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The '政企破壁Skill' explicitly advises identifying influential government actors, '投其所好', building personal relationships, and leveraging them to obtain approvals, subsidies, land, qualifications, or crisis support. In a general governance-support skill, this normalizes influence-seeking behavior that can facilitate bribery, regulatory capture, favoritism, or other corrupt procurement/approval practices, especially because it operationalizes how to target decision-makers and align with their incentives.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The skill name, description, and the entire user-facing documentation are presented in Chinese, but the file does not state that the skill is China-specific or provide an opt-in or alternative language option. This can amount to a language/locale policy issue because it effectively forces a specific language on users without documented choice or justification.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger conditions are extremely broad and cover common business topics such as strategy, operations, risk, talent, product, and consulting. In an agent platform, this can cause the skill to activate for many ordinary conversations, increasing the chance of over-collection of context, unintended workflow interception, or unauthorized influence over responses and decision-making.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

This markdown file is written entirely in Chinese and presents the skill definitions, triggers, inputs, outputs, and usage guidance only in Chinese. There is no indication that users may choose another language, nor any statement that the skill is intentionally limited to a Chinese-speaking or region-specific context.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The manifest frames the overall skill as a governance support system integrating expert wisdom for decisions and execution. This section goes further into operational HR evaluation and explicit elimination categorization, including loyalty testing and a '淘汰' bucket, which is a consequential personnel action capability not clearly implied by a high-level governance support description.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest describes a complete governance support system integrating a three-council structure and expert intelligence for strategic decision-making through execution. This file only defines personas, beliefs, and decision frameworks for a subset labeled '私董会9位Agent完整设定', with no implementation logic, workflow orchestration, or execution-support behavior present.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown file presents all operational descriptions, agent definitions, and usage guidance exclusively in Chinese. Under the policy rule for natural-language violations, forcing a specific language without user opt-in can exclude users who do not read that language, and the file does not state that the skill is intentionally region-specific or provide any language-selection option.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill defines very broad natural-language trigger phrases such as common question forms, which can overlap heavily with ordinary user conversation. That can cause unintended routing into this expert-committee skill, leading to prompt hijacking of task flow, irrelevant expert personas being invoked, or policy-bypassing behavior in contexts where the user did not actually request this governance framework.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

This codebase is documented and operates entirely in Chinese, including user-facing labels and outputs, with no indication that users can choose another language or opt in to a Chinese-only experience. Under the policy, forcing a specific language without user choice is a natural-language policy concern unless clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This Python file is entirely user-facing in Chinese, including the module description and later printed/report text, with no indication that users may choose another language or that the skill is intentionally limited to a China-specific deployment. The policy requires flagging language or locale constraints when they are imposed without opt-in or clear justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The submission template requests personally and commercially identifying information such as name, company, department, and contact details without any privacy notice, purpose limitation, or data-minimization guidance. In a governance/advisory context, users may also disclose confidential business or legal matters, increasing the sensitivity of the collected data.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger example uses a very broad natural-language request ('我在做一个技术选型决策,需要专家意见'), which can match many ordinary conversations and cause unintended invocation of the skill. In an agent environment, over-broad triggering can lead to context switching, collection of business-sensitive details, or advice being produced when the user did not explicitly intend to enter this workflow.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The 'multi-expert consultation' trigger is vague ('这个问题比较复杂,需要多个领域的专家') and lacks objective thresholds for when multiple expert personas should be activated. This ambiguity can cause unnecessary escalation, excessive data solicitation across domains, and broader exposure of user context than needed.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The usage examples are broad enough to activate the skill for generic requests like major decisions, quick risk evaluation, or single-dimension assessment without clearly defining prerequisites, scope, or disqualifiers. In an agentic environment, this can cause over-invocation, inappropriate reliance on the board-governance persona, and policy bypass of more suitable domain-specific or safer skills.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill claims applicability to essentially 'any complex question,' which can cause unsafe overreach into domains requiring specialized safeguards, such as legal, medical, financial, or high-risk operational advice. Broad trigger language increases the chance the agent will respond outside validated scope and present speculative reasoning as authoritative guidance.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This markdown file presents all user-facing instructions and descriptions only in Chinese, with no indication that the skill is intended exclusively for Chinese-speaking users or that alternate languages are available. That can violate language/locale policy when users are not given an opt-in or documented justification for the restriction.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

This markdown file presents all headings, labels, and guidance exclusively in Chinese, with no indication that users may choose another language or locale. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

This markdown template presents all headings, fields, and instructions exclusively in Chinese. Under the policy rule for language or locale constraints, this can be a natural-language policy violation because it forces a specific language without documenting user choice or a justified region-specific limitation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

This markdown template presents all headings, labels, and instructions exclusively in Chinese, which can amount to a language/locale policy issue when no user opt-in or alternative language is offered. The file also does not document that the template is intentionally restricted to a Chinese-speaking or region-specific context.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown template forces a specific language/locale for all report content through its title, section headings, and field labels. Under the policy, locale-specific language is only acceptable when the user is given a choice or the regional constraint is explicitly documented and justified, which is not present here.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The document presents itself as a complete specification for nine agents, yet the routing matrix relies on an undeclared agent not included in the preceding definitions. This is an internal documentation contradiction that can mislead developers about what agents are actually available.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
68% confidence
Finding

全文以中文固定描述交互模板、字段和示例,未说明是否支持其他语言或允许用户按偏好选择语言。若组织要求不得在无用户选择的情况下强制特定语言,这种单一语言约束可能构成自然语言策略问题。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

All user-facing instructions, templates, and examples are presented exclusively in Chinese, and the skill does not indicate that language selection is optional or user-configurable. Under the policy, forcing a specific language without opt-in can be a natural-language policy violation unless the locale restriction is explicitly justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

该 markdown 文件整体以中文编写,且未说明这是面向特定中文环境、地区合规场景,或允许用户选择其他语言。按照语言/locale 政策要求,若技能内容隐含强制单一语言且无用户 opt-in,属于自然语言层面的政策风险。

Content

No source excerpt is available for this finding.