Security audit

Rstudio Runner

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed bioinformatics workflow assistant that writes local analysis outputs, with no bundled executable code, credential use, exfiltration, or hidden behavior found.

Install this only if you want an agent to generate and run local R/RStudio bioinformatics workflows. Use a dedicated project folder, confirm where outputs will be written, review generated R scripts before execution, and treat reports, logs, and RDS files as potentially sensitive biomedical data.

SkillSpector

Vulnerability Patterns

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

Findings (5)

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill registers broad triggers such as 'bioinformatics', 'run analysis', and '自动分析', which can cause the skill to activate in contexts where the user did not intend filesystem changes or automated analysis. Because the skill advertises automatic script generation, execution, and file organization, unintended invocation could lead to unwanted processing, data exposure in outputs, or disruptive workspace modifications.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill repeatedly promises automatic folder creation, script writing, output saving, and project reorganization, but does not describe a mandatory warning or confirmation before modifying the filesystem. In a skill that may trigger on generic keywords and common data file extensions, silent writes increase the risk of overwriting files, cluttering directories, or placing sensitive analysis outputs in unintended locations.

Natural-Language Policy Violations

Medium

Confidence: 79% confidence
Finding: Mandating Chinese report generation without opt-in is primarily a safety and usability issue rather than a classic security flaw, but it can still have security implications if users cannot easily review generated findings, warnings, or data-handling behavior. In a system that automates bioinformatics workflows and produces many artifacts, lack of language choice may reduce informed consent and hinder validation of sensitive output content.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The trigger set is broad and generic enough to activate on common bioinformatics and analysis-related terms such as 'bioinformatics', 'run analysis', and generic file types like .csv, .txt, and .html. In a skill that can auto-organize files, generate scripts, and run analysis tools, unintended activation could cause unrequested processing, file modification, or execution workflows on unrelated user data.

Natural-Language Policy Violations

Medium

Confidence: 77% confidence
Finding: The description advertises Chinese report generation as a built-in behavior without indicating that the output language is configurable or user-selected. This can lead to unwanted disclosure, confusion, or operational errors if reports are automatically produced in a language the user or downstream reviewers do not understand, especially in scientific or clinical-adjacent workflows.

Static analysis

No suspicious patterns detected.