Milvus Skill

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Milvus database helper skill, with normal database-admin risks that users should control carefully.

Install only if you intend to let your agent help manage Milvus resources. Treat delete, drop, truncate, password, role, and privilege examples as real administrative actions: provide least-privilege credentials, verify the target database or collection, and require explicit confirmation before destructive operations.

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: 87% confidence
Finding: The documentation includes a direct example of `client.drop_database(db_name="my_database")` without any warning, confirmation step, or guidance about validating user intent before performing this irreversible destructive action. In an agent skill context, examples often become implementation templates, so this increases the chance an agent or developer will automate database deletion in response to ambiguous or malicious prompts, causing data loss.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal