Security audit

HR面试评价助手-Word版

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple HR interview template, but it asks users to reproduce sensitive candidate details that may be unnecessary for hiring evaluation.

Review this skill carefully before installing for real hiring workflows. Use it only with authorized candidate data, remove or redact sensitive personal fields that are not job-related, and avoid using protected characteristics such as sex, marital/fertility status, family circumstances, or detailed home address in hiring decisions or shared evaluation documents.

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

Findings (1)

other

Warning

Location: SKILL.md:32
Finding: Excessive Collection and Plaintext Reproduction of Candidate Personal Data
Content: View full analysis
Remediation: View remediation

SkillSpector

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 (3)

Missing User Warnings

High

Category: Not specified by scanner
Confidence: 96% confidence
Finding: This skill explicitly processes resumes, interview notes, salary details, family status, home location, and other candidate personal data, but it provides no warning or handling guidance for sensitive information. In an HR context, this raises real privacy and compliance risks because users may upload highly sensitive PII/employee data without minimization, consent checks, or redaction guidance.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The skill description says users can invoke it by sending the phrase “HR面试评价”, but it does not define any scope limits, exclusion conditions, or negative examples. While somewhat domain-specific, the manifest presents it as a free-form口令 without clarifying when the skill should or should not activate, which can lead to unintended invocation in ordinary HR discussion.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The manifest description and invocation instructions are entirely in Chinese and present the skill as producing a Chinese HR interview evaluation format, with no indication that users may choose another language or locale. Under the policy, a fixed language output without opt-in or documented justification is a natural-language policy concern.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.