ClawHub
Back to skill

Security audit

上市公司违规案例周报

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent reporting tool that queries a local violation-case database and renders weekly reports without evidence of hidden data access, mutation, persistence, or exfiltration.

Before installing, confirm you expect this skill to contact the local weigui MCP service and generate reports from that database. Its implicit invocation setting means it may be selected automatically for matching violation-report requests, but the inspected artifacts do not show hidden network exfiltration, data mutation, or persistence.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (1)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The skill enables allow_implicit_invocation without any visible narrowing conditions, so the agent may auto-select this skill based on broad matching rather than explicit user consent. Because the skill reaches an MCP service over HTTP and can generate deliverable outputs, unintended invocation could cause unauthorized data retrieval, surprise tool use, or disclosure of sensitive case-report content in contexts where the user did not intend this skill to run.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.install_untrusted_source

Install source points to URL shortener or raw IP.

Warn
Code
suspicious.install_untrusted_source
Location
agents/openai.yaml:11