Baidu Milan Winter Olympics 2026

• Outcome: Pass
• Latest audit: May 28, 2026 at 10:52 AM UTC
• Version:
  0.0.2

Overview

This skill fetches public Winter Olympics data from Baidu Sports and does not show credential use, local data access, persistence, or destructive behavior.

Install only if you are comfortable running local Node.js scripts that scrape Baidu Sports. The main practical concern is the randomized browser User-Agent behavior; compliance-sensitive users should prefer a transparent client identifier, rate limits, and confirmation that this access is allowed by the source site.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Context-Inappropriate Capability

Medium

• Confidence: 89% confidence
• Finding: The script rotates through a pool of 20 User-Agent strings and explicitly states this is to avoid using a fixed UA, which is an anti-detection/evasion technique rather than a functional requirement for simple data retrieval. In a scraping skill, this increases concern because it facilitates bypassing source-site rate limits, fingerprinting, or bot-detection controls and can enable repeated unauthorized collection at scale.

Context-Inappropriate Capability

Medium

• Confidence: 93% confidence
• Finding: The script deliberately rotates among many realistic browser User-Agent strings and comments that this is to avoid using a fixed UA while scraping Baidu. That behavior is not required to fulfill the stated business function of retrieving Olympic schedules and materially increases stealth/evasion characteristics, which can facilitate bypassing anti-bot controls or terms-based restrictions on the upstream service.

Intent-Code Divergence

Low

• Confidence: 82% confidence
• Finding: The header comment frames the code as simply consuming an asynchronous API, but the implementation includes browser impersonation and randomization intended to make requests look like different end-user clients. This mismatch is a security-relevant transparency issue because it hides evasive collection behavior behind innocuous documentation, making review and governance harder.

VirusTotal

66/66 vendors flagged this skill as clean.