ClawHub

Vector DB Toolkit

Security checks across malware telemetry and agentic risk

Overview

This is a coherent vector database toolkit with expected database, persistence, and optional embedding API behavior, but users should handle sensitive text and dependency versions carefully.

Install only if you need vector database/RAG utilities. Use test collections first, verify collection names before delete calls, choose an intentional Chroma storage path, and do not use provider='openai' with confidential text unless sending that text to OpenAI under your API key is acceptable. For production, install with pinned, scanned dependencies.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill documentation and examples indicate capabilities that use network access and potentially environment-derived configuration, but the skill declares no permissions. That mismatch is a real security issue because agents or reviewers may underestimate what the skill can do, especially when it connects to external vector databases or embedding providers. In this context, the toolkit is legitimately intended to access services like Qdrant/Chroma, so the issue is more about transparency and policy enforcement than obviously malicious behavior.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The code sends caller-provided text to a third-party API and uses a credential to do so, but it provides no disclosure, consent boundary, or sensitivity checks. In AI/RAG contexts, those texts may contain proprietary data, user prompts, or secrets, so silent transmission can create a real privacy and compliance risk even if the behavior is functionally intended.

Unpinned Dependencies

Low
Category
Supply Chain
Content
qdrant-client>=1.7.0
chromadb>=0.4.18
numpy>=1.24.0
requests>=2.31.0
Confidence
94% confidence
Finding
qdrant-client>=1.7.0

Unpinned Dependencies

Low
Category
Supply Chain
Content
qdrant-client>=1.7.0
chromadb>=0.4.18
numpy>=1.24.0
requests>=2.31.0
sentence-transformers>=2.2.2
Confidence
93% confidence
Finding
chromadb>=0.4.18

Unpinned Dependencies

Low
Category
Supply Chain
Content
qdrant-client>=1.7.0
chromadb>=0.4.18
numpy>=1.24.0
requests>=2.31.0
sentence-transformers>=2.2.2
Confidence
90% confidence
Finding
numpy>=1.24.0

Unpinned Dependencies

Low
Category
Supply Chain
Content
qdrant-client>=1.7.0
chromadb>=0.4.18
numpy>=1.24.0
requests>=2.31.0
sentence-transformers>=2.2.2
Confidence
95% confidence
Finding
requests>=2.31.0

Unpinned Dependencies

Low
Category
Supply Chain
Content
chromadb>=0.4.18
numpy>=1.24.0
requests>=2.31.0
sentence-transformers>=2.2.2
Confidence
92% confidence
Finding
sentence-transformers>=2.2.2

Known Vulnerable Dependency: qdrant-client — 1 advisory(ies): CVE-2024-3829 (qdrant input validation failure )

Critical
Category
Supply Chain
Confidence
88% confidence
Finding
qdrant-client

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal