Security audit

Scholarship

Security checks for vulnerabilities and agentic risk

Overview

The skill claims to help with scholarships but its actual instructions are about paid courses and learning-platform features, so it should be reviewed before use.

Review this skill before installing. It does not appear to contain code or request system access, but users expecting scholarship guidance may instead receive course or paid-learning recommendations. The publisher should align the name, description, examples, and body before release.

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

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 (4)

Intent-Code Divergence

High

Confidence: 98% confidence
Finding: The manifest/title claim a scholarship assistant, but the body documents a different tool focused on learning products and course operations. This contradiction undermines trust boundaries and can cause incorrect invocation, user deception, or policy evasion by disguising one capability as another.

Description-Behavior Mismatch

High

Confidence: 98% confidence
Finding: The documented behavior does not match the declared purpose of a scholarship guidance skill. Instead of scholarship rules, deadlines, and application help, it presents course-selection and paid-learning features, which can misroute users, trigger the wrong skill, or cause the agent to provide irrelevant or commercially biased guidance in a sensitive academic/financial context.

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: The description and all user-facing instructions are written entirely in Chinese, and the skill does not indicate that users may interact in other languages or opt into this locale. Under the stated policy, a language or locale constraint should either be optional for the user or clearly justified as region-specific.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The example trigger phrases are extremely generic and can overlap with ordinary help-seeking requests, increasing the chance of unintended routing or activation. In combination with the skill’s mismatched identity, this broad phrasing makes accidental invocation more dangerous because users asking for unrelated assistance may receive inappropriate scholarship/course guidance.

Static analysis

No suspicious patterns detected.