ClawHub

Hugging Face

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a transparent Hugging Face helper that only sends user-selected searches and prompts to Hugging Face services.

Install only if you are comfortable sending selected prompts, search terms, and inference inputs to Hugging Face. Use a scoped HF_TOKEN, do not include secrets or sensitive private data in prompts, and review any local ~/hugging-face memory notes before relying on them.

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 (1)

External Transmission

Medium
Category
Data Exfiltration
Content
```bash
MODEL_ID="google/flan-t5-large"

curl -s "https://api-inference.huggingface.co/models/${MODEL_ID}" \
  -H "Authorization: Bearer ${HF_TOKEN}" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
60% confidence
Finding
curl -s "https://api-inference.huggingface.co/models/${MODEL_ID}" \ -H "Authorization: Bearer ${HF_TOKEN}" \ -H "Content-Type: application/json" \ -d

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal