Security audit

Genome Report

Security checks across malware telemetry and agentic risk

Overview

Review before installing: this is a local genome-reporting tool, but it processes and saves very sensitive genetic and family information without strong privacy or consent guardrails.

Install only if you intentionally want local analysis of genome files. Use data only with the person's informed consent, avoid family comparison unless every included person agrees, choose private non-synced storage for outputs, and delete or protect generated HTML/JSON reports because they may contain durable identifying health and trait information.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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 (6)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 84% confidence
Finding: The skill documentation advertises HTML report generation with an output path, which implies file creation/write behavior, but no permissions are declared. Undeclared write capability weakens reviewability and user consent boundaries, especially for sensitive genome-derived reports that may be written to disk in predictable or user-chosen locations.

Vague Triggers

Medium

Confidence: 78% confidence
Finding: The trigger language is broad enough to activate on general requests about DNA, genetic reports, health risk, or family comparison without clearly signaling the sensitivity of the task. That increases the chance the skill is invoked on highly sensitive genetic data in contexts where the user did not fully understand that raw genome processing or family-member comparison would occur.

Missing User Warnings

High

Confidence: 95% confidence
Finding: The markdown omits a clear privacy warning despite handling raw genetic data and family comparison workflows, both of which are extremely sensitive and potentially identifying. Without prominent disclosure, users may expose health-risk indicators, ancestry information, pharmacogenomic traits, and relatives' data without understanding the privacy, consent, and retention implications.

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: This database includes ethnicity-linked and socially sensitive trait labels such as skin pigmentation, body odor, and ancestry-associated descriptors without medical, scientific, or ethical guardrails. In a genome-reporting skill, presenting such traits as simple genotype-to-identity mappings can enable stereotyping, stigmatization, or overconfident inferences about race, appearance, or social characteristics from DNA data.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The script writes an HTML report containing highly sensitive genome-derived health and trait data to disk by default naming convention, but it does not provide any explicit warning or consent gate about local persistence, file sensitivity, or safe storage. In this skill context, genetic data is exceptionally sensitive, and saving it to an unprotected HTML file can expose medical-risk inferences and personal traits to other local users, backups, sync services, or accidental sharing.

Missing User Warnings

High

Confidence: 97% confidence
Finding: The family comparison path saves aggregated multi-person genetic results to disk, increasing sensitivity because it combines health-related inferences across relatives in one file. This materially raises privacy risk: the persisted report can reveal shared hereditary risks and family relationships, making accidental disclosure or unauthorized access more damaging than a single-user report.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.