ClawHub
Back to skill

Security audit

知识库健康度检测

Security checks across malware telemetry and agentic risk

Overview

This skill reads user-provided FAQ and support-dialogue files to produce a comparison report, with no evidence of hidden persistence, network transfer, or unrelated actions.

Before installing, use this only with FAQ and dialogue files you intend the agent to read. Remove customer personal information where possible, and be aware the helper prints file contents into the analysis context.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding
The skill requires reading user-supplied files via `scripts/read_file.py`, but it does not declare any permissions for file access. Undeclared file-read capability is dangerous because it obscures the skill’s real access requirements and can enable broader-than-expected ingestion of local file contents if the helper script accepts arbitrary paths.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding
The documented purpose is FAQ/dialogue comparison, but the behavior described includes generic reading and textual expansion of multiple file types, including arbitrary allowed-path files and `.json`, which is not disclosed in inputs. This mismatch is dangerous because it can mask a broader data-extraction capability than users expect, increasing the risk of unintended sensitive-data exposure through helper-script file access.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.