Back to skill

Security audit

Needle

Security checks across malware telemetry and agentic risk

Overview

This skill is a clearly disclosed Needle connector that can read and create Needle content through the OOMOL CLI, with no hidden executable payloads found.

Install only if you intend the agent to use your connected OOMOL/Needle account. Review write actions before approval, especially creating collections or adding files for indexing, because those can change what Needle stores and searches.

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
90% confidence
Finding
The trigger phrase is overly broad: 'Use this skill for ANY Needle request' and the description can cause the skill to be invoked for any task merely mentioning Needle, even when direct API use or another skill would be more appropriate. This increases the chance of unintended tool execution and expands the attack surface, especially in agentic systems where invocation routing is automated.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.