hash health

Security checks across malware telemetry and agentic risk

Overview

This Hash Health skill is coherent, but it can immediately send and save sensitive meal and health data to a remote account with broad triggers and limited confirmation.

Install only if you trust the Hash Health hosted service with food photos, meal history, medication details, and your HASH_HEALTH_TOKEN. Use explicit wording when you want data saved, and avoid invoking it during casual food or health discussion if you do not want information sent or logged.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The skill explicitly instructs the agent to upload a user's food image to a remote service immediately, before any user-facing notice or confirmation. Images can contain sensitive health, location, or incidental personal information, so transmitting them automatically creates a privacy risk and weakens meaningful user consent.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill tells the agent to immediately send meal text such as 'I ate X' or 'log X' to the remote service without first warning the user that their dietary data is being transmitted and stored. Dietary intake can reveal health conditions, religious practices, and lifestyle patterns, so silent transmission increases privacy and consent risk.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal