ClawHub

LiteLLM

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward LiteLLM helper that sends user-provided prompts to configured LLM providers or a proxy, with no evidence of hidden behavior.

Install only if you are comfortable with prompts, system messages, and related metadata being sent to the selected LLM provider or LiteLLM proxy. Avoid sending secrets, regulated data, or confidential code unless that provider or proxy is approved for it, protect API keys, monitor usage costs, and consider pinning the LiteLLM dependency in managed environments.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (2)

Missing User Warnings

Medium
Confidence
96% confidence
Finding
This skill is explicitly designed to send prompts to external LLM providers and possibly a LiteLLM proxy, yet the description does not warn users that their prompts and other content may leave the local/runtime boundary. Without a clear disclosure, users may provide sensitive source code, credentials, or proprietary data under the false assumption that processing remains internal.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
This helper sends the supplied prompt and optional system message to an external LLM provider via LiteLLM, but it gives no warning, confirmation, or guardrails about network transmission or potential disclosure of sensitive data. In a skill context, users may paste secrets, source code, customer data, or internal prompts, making silent exfiltration to third-party services a real privacy and compliance risk.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal