Security audit

Career Planner 职业规划顾问

Security checks for vulnerabilities and agentic risk

Overview

This career-planning skill is coherent and not malicious, but it should be reviewed because it broadly triggers, asks for sensitive career and income details, and always saves a personalized report without explicit consent.

Install only if you are comfortable sharing career history, skills, income goals, location/work preferences, and constraints for a saved career report. Users should be offered clear career-intent confirmation and explicit consent before saving the report, especially when the request is vague or contains personal financial details.

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
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The trigger criteria are excessively broad and include vague everyday phrases like 'I don't know what to do' and 'I want a change', which can cause the skill to activate outside genuine career-planning contexts. This creates overreach risk: the agent may collect unnecessary personal data, steer conversations away from user intent, or invoke file-writing/web-research behaviors when the user did not ask for a career workflow.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The instructions say to "Write in Chinese (unless user clearly prefers English)" and later to "Default to Chinese (Mandarin) unless the user writes in English." This imposes a default language choice rather than first offering the user a language preference, which can violate language/locale choice expectations.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The skill mandates writing a personalized report to persistent user storage and presenting it for download without first obtaining explicit consent or warning the user that their data will be saved. Because the report is based on a detailed intake covering employment status, income goals, constraints, and other potentially sensitive personal information, this can create privacy and data-retention risks if users do not expect persistent storage.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The template hard-codes Chinese report output regardless of the user's language preference or locale, which can cause the agent to disclose advice in an unexpected language and degrade usability, informed consent, and accessibility. In a career-planning skill, this is especially risky because recommendations, salary notes, and action steps may be misunderstood or unusable if the user did not request Chinese output.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The instruction to use Chinese terms for a report field imposes a specific language/locale convention in the output. The file does not indicate that this is optional, user-selected, or justified as a region-specific requirement, which can violate language choice policy.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 81% confidence
Finding: This markdown reference is presented as a general skill-to-role mapping resource, but substantial user-facing content is in Chinese while surrounding instructions are in English. Because the file does not state that it is intended only for Chinese-speaking users or provide any language-choice/opt-in mechanism, it may violate language/locale policy expectations.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.