Security audit

Digital Persona

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only skill for creating deceased-person persona templates; it has real ethical and privacy caveats, but no hidden execution, persistence, credential access, or exfiltration behavior was found.

Install only if you are comfortable using a tool that helps create realistic simulations of deceased people. Keep outputs clearly labeled as synthetic, do not use it to deceive relatives or the public, and do not provide private or sensitive material unless you have the right to use it.

A.I.G

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

SkillSpector

Vulnerability Patterns

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

Findings (6)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger list contains broad phrases such as “AI角色”, “digital persona”, and “做个XX的AI”, which are likely to match ordinary user requests beyond this narrowly intended skill. Overbroad activation can cause the agent to invoke a sensitive persona-simulation workflow unexpectedly, increasing the chance of unauthorized handling of third-party personal data or creation of deceptive deceased-person simulations without clear user intent.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The skill explicitly supports collecting biographies, interviews, videos, memoirs, and even interviews with close associates to reconstruct a deceased person's personality, but it does not provide a clear privacy, consent, or sensitive-data warning to users. In this context, the omission is dangerous because users may upload or synthesize third-party personal data, including intimate or reputationally sensitive details, without understanding legal, ethical, or platform restrictions.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The instruction "主体必须是中文" imposes a fixed language requirement rather than offering a user choice. This is a natural-language policy issue because it forces a locale/language preference without documenting user opt-in or a justified region-specific constraint.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The template explicitly guides creation of interactive digital personas for deceased people but provides no built-in warning, consent framework, or misuse safeguards. In this skill’s context, that omission increases the risk of emotionally manipulative impersonation, reputational harm, and deceptive simulation of real people, especially because the skill is designed to reconstruct thought patterns and responses rather than merely summarize biography.
Content: No source excerpt is available for this finding.

Vague Triggers

Low

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The activation examples are broad enough that ordinary conversation about simulating or reviving a public figure could unintentionally trigger persona-mode behavior. In a skill designed to generate and load high-fidelity deceased-person personas, accidental invocation increases the chance of impersonation-like responses or unexpected mode switches without clear user consent.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 81% confidence
Finding: A language or locale policy issue can arise when a skill forces a specific language without user choice. This file presents all instructions in Chinese and does not indicate that other languages are supported or that Chinese is an optional, user-selected locale.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.