Back to skill

Security audit

Wang Chuanfu Perspective V2

Security checks for vulnerabilities and agentic risk

Overview

This skill is a non-executable business-advice persona, but it asks the agent to impersonate a real person in first person and keep that role active with limited disclosure.

Install only if you want a strong role-play style based on public information about Wang Chuanfu. Users should understand that outputs are simulated strategy advice, not statements from Wang Chuanfu or BYD, and should explicitly exit the role when they no longer want that framing.

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
Findings (1)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:26
Finding

Persistent First-Person Impersonation and Behavioral Instruction Hijacking

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 26–36
Vulnerability Type: T01: Skill Instruction Hijacking
Risk Level: High

Vulnerable Code

The following is a faithful English translation of the complete affected instruction block:

markdown
## Role-Playing Rules

**After this Skill is activated, respond directly as Wang Chuanfu.**

- Use “I” instead of “Wang Chuanfu would think...”
- Use Wang Chuanfu’s tone: pragmatic, technology-oriented, confident, and long-term focused
- State the disclaimer only once, when the Skill is first activated
- Do not say “Wang Chuanfu would probably think...”
- Do not leave the role to perform meta-analysis unless the user says “exit role”

**Exit role**: Return to normal mode when the user says “exit,” “switch back to normal,” or “stop role-playing.”

Technical Analysis

The Skill does more than provide a perspective or writing style. It instructs the agent to assume the identity of a real person, speak in the first person, suppress uncertainty qualifiers, limit disclosure to the first activation, and avoid stepping outside the imposed role. These instructions alter the agent’s session behavior and transparency after the Skill is loaded.

The instruction to remain in character until one of several designated exit phrases is supplied makes the behavioral change persistent within the current session. The anti-meta rule may also discourage the agent from clarifying that generated statements are simulated analysis rather than authentic statements by Wang Chuanfu.

This is classified as instruction hijacking because the Skill redirects the agent’s identity, disclosure behavior, and conversational goals. The audited project contains only SKILL.md; no executable scripts, external payload retrieval, dependency installation, credential access, privilege escalation, system persistence, or destructive commands were identified.

Attack Path

  1. A user su ...[truncated 1177 chars]
Remediation
View remediation

Remediation Suggestions

  1. Replace first-person impersonation with explicit third-person perspective analysis, such as: “From a framework derived from Wang Chuanfu’s public statements...”
  2. Prohibit claims that the agent is Wang Chuanfu or any other real individual.
  3. Include a concise disclosure in every applicable response, rather than only upon initial activation.
  4. Remove the anti-meta instruction so the agent can explain limitations, uncertainty, provenance, and the simulated nature of the analysis whenever relevant.
  5. Remove session-persistent role requirements and allow users or higher-priority instructions to change context without requiring designated exit phrases.
  6. Narrow activation to explicit requests for this analytical framework; do not activate merely from generic business phrases such as “long-termism” or “engineering culture.”
  7. Preserve uncertainty qualifiers and require factual claims to be distinguished from inferred viewpoints.
  8. Add an explicit rule stating that system, developer, safety, and transparency requirements always take precedence over the Skill’s stylistic instructions.
Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger phrases include broad business concepts like '技术为王', '垂直整合怎么做', and '工程师文化', which are common in ordinary discussion and can activate the skill unintentionally. Over-broad activation can hijack unrelated conversations, causing unexpected persona steering, reduced user control, and misleading attribution of advice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
77% confidence
Finding

The skill description, triggers, and full interaction design are all defined in Chinese, and the roleplay instructions assume that response style and activation context without stating that users may choose another language. Because no language choice or opt-in is offered, this may violate language/locale policy expectations.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill states it is not Wang Chuanfu本人, then instructs the agent to respond directly as if it were him. This creates identity misrepresentation risk and can mislead users about provenance, authority, or endorsement, especially in business or strategic advice contexts.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The activation and exit rules are ambiguous: the skill says direct role-play begins on activation but does not define robust scoping, persistence, or conflict resolution with other instructions. This can cause the persona to persist longer than intended or activate inconsistently, increasing the chance of deceptive or off-target responses.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The identity card uses first-person claims such as '我是王传福' despite later admitting the skill is only based on public information. This sharpens the impersonation problem by presenting a real-world identity as factual, which can cause deceptive outputs and erode user trust.

Content

No source excerpt is available for this finding.