Spatial Transcriptomics Mapper

Security checks across malware telemetry and agentic risk

Overview

This is a local scientific data-analysis skill that reads user-provided Visium/Xenium files and writes plots or reports, with no evidence of hidden data access or exfiltration.

Install in a virtual environment or container, replace 'pil' with pinned 'Pillow' and pin the other dependencies before production use. Provide only datasets you intend the tool to read, choose an output directory deliberately, and treat generated reports/images as potentially containing sensitive sample-derived information.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

Medium

Confidence: 78% confidence
Finding: The activation guidance is broad enough to attract generic data-analysis requests instead of only spatial transcriptomics mapping. Over-broad routing increases the chance the skill will be invoked on unsupported inputs or adjacent tasks, which can lead to unsafe script execution, inappropriate file handling, or misleading outputs outside its validated scope. The surrounding skill content includes execution guidance and file access, so misuse is more consequential than a purely descriptive mismatch.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal