Codereadr

Security checks across malware telemetry and agentic risk

Overview

This appears to be a real CodeREADr/Membrane helper, but it gives an agent broad account-changing and delete abilities without clear safety checks.

Install only if you intend to let an agent operate on a real CodeREADr account through Membrane. Use the least-privileged account or connection available, verify the Membrane CLI source/version, and require explicit confirmation before any create, edit, delete, billing, role, permission, API-key, or raw API request.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill advertises destructive actions like deleting scans, databases, users, and services without any warning, safeguard language, or confirmation requirement. In an agent setting, this increases the risk that a vague or mistaken user request could result in irreversible data loss in a live CodeREADr environment.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal