ClawHub

rrragflow-skill

Security checks across malware telemetry and agentic risk

Overview

This skill coherently manages RAGFlow datasets and documents using a user-configured API key and endpoint, with disclosed upload, search, update, parsing, and deletion capabilities.

Install only if you trust the publisher and the configured RAGFlow server. Use a least-privileged RAGFlow API key, avoid uploading confidential or regulated files unless authorized, and carefully confirm exact dataset or document IDs before any update, delete, parse, or stop action.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill clearly uses sensitive capabilities via environment variables, local file paths for uploads, and network access to a remote RAGFlow API, yet it does not declare explicit permissions beyond runtime requirements. This weakens policy enforcement and user awareness because the agent can access secrets and exfiltrate local file contents to an external service without a formal permission boundary.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill advertises dataset and document deletion capabilities but does not present a prominent general warning in the description about irreversible data loss. Although later guardrails require confirmation, the top-level description still understates destructive risk, which can lead to accidental invocation in a high-impact data management context.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal