Football Prediction Optimized

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:42 PM UTC
• Version:
  1.0.0

Overview

This football prediction skill behaves consistently with its purpose, using user-supplied match text and a disclosed DeepSeek API key to generate predictions.

Install only if you are comfortable sending match feature text to DeepSeek or the configured DeepSeek-compatible endpoint. Use a dedicated API key, verify DEEPSEEK_BASE_URL before running, avoid private/proprietary notes in input files, and review predictions carefully before using them for betting or other financial decisions.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 92% confidence
• Finding: The skill advertises executable behaviors that use environment variables, local file access, and network access, but it does not declare permissions or surface these capabilities clearly in a permission model. This weakens user consent and platform enforcement, making it easier for the skill to read local inputs and send data externally without adequate transparency.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 96% confidence
• Finding: The skill’s stated purpose focuses on football prediction, but the implementation also sends user-provided match data to the external DeepSeek API, which is a material behavior not prominently disclosed in the high-level description. Hidden or underemphasized third-party transmission creates privacy, compliance, and trust risks, especially if users assume processing is local or confined to the skill ecosystem.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger examples are broad enough that the skill may auto-activate for generic requests about match analysis or prediction without clear boundaries. Over-broad activation can cause unintended execution, including accidental file processing or network calls to third-party services when the user did not explicitly request this specific skill.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill documents use of a required API key and an external DeepSeek endpoint but does not provide a user-facing privacy or data-transfer warning. Users may submit match feature text or other contextual data without understanding that it will be sent to a third party, creating avoidable confidentiality and compliance risk.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill reads arbitrary match input from a local file and sends the full contents to the external DeepSeek API for processing, but the code provides no user-facing notice, consent step, redaction, or data-classification guardrails before transmission. This creates a real data exfiltration/privacy risk if the input file contains proprietary betting data, personal information, or other sensitive material beyond what the user expected to leave the local environment.

VirusTotal

67/67 vendors flagged this skill as clean.