Security audit

Alumni Career Tracker

Security checks across malware telemetry and agentic risk

Overview

This is a local alumni career analytics tool that handles personal alumni data, so privacy practices matter but the artifacts do not show hidden collection, exfiltration, or unsafe automation.

Install only if you are prepared to handle alumni personal data responsibly. Obtain consent, avoid collecting salary or LinkedIn details unless necessary, keep exports private, share only aggregate reports publicly, and consider pinning dependencies or using an isolated Python environment.

SkillSpector

By NVIDIA

Vulnerability Patterns

Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (5)

Context-Inappropriate Capability

Medium

Confidence: 88% confidence
Finding: The documented `--update-linkedin` and LinkedIn monitoring behavior expands data collection beyond core alumni analytics into third-party profile tracking, creating privacy and consent risk. In this context, scraping or refreshing personal employment data from LinkedIn can collect, process, or infer sensitive professional information without a clearly stated lawful basis or user consent workflow.

Intent-Code Divergence

High

Confidence: 95% confidence
Finding: The skill claims reports should be anonymized and aggregate-only, yet the examples include exporting identifiable alumni contacts for outreach. This creates a real risk of unauthorized disclosure of personal data and function creep from analytics into targeted contact list generation, which is especially sensitive in an academic mentoring context.

Unpinned Dependencies

Low

Category: Supply Chain
Content: dataclasses pandas rich
Confidence: 92% confidence
Finding: The dependency is unpinned, which allows future installs to resolve to different versions over time. This creates supply-chain and reproducibility risk because a later version could introduce breaking changes or a vulnerable release without any code changes in the skill.

Unpinned Dependencies

Low

Category: Supply Chain
Content: dataclasses pandas rich
Confidence: 98% confidence
Finding: The unpinned pandas dependency means installations may pull any available version, including versions with security issues or incompatible behavior. In a data-analysis skill, pandas is central functionality, so lack of version control increases both supply-chain exposure and the chance of unintentionally installing an unsafe release.

Unpinned Dependencies

Low

Category: Supply Chain
Content: dataclasses pandas rich
Confidence: 91% confidence
Finding: The rich package is also unpinned, so dependency resolution is nondeterministic and may fetch a later compromised or vulnerable version. While rich is typically used for console formatting and is lower risk than core data-processing libraries, it still expands the supply-chain attack surface.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.