ClawHub
Back to skill

Security audit

BidHunter 标讯猎手

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent procurement-monitoring tool that fetches public tender notices, evaluates them with local rules, and writes local reports.

Before installing, be aware that running the pipeline contacts public procurement websites and stores fetched notices plus generated reports locally. Configure qual_rules.json with your own business rules, confirm any scheduled run or IM/email sharing yourself, and avoid opening or forwarding generated HTML reports from untrusted source data without sanitizing them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill clearly instructs users to run shell scripts, read and write local files, and generate reports, yet it does not declare corresponding permissions. Undeclared capabilities reduce transparency and can bypass normal user expectations or policy checks, especially because the workflow includes filesystem access and command execution.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The trigger phrases are broad and map to common user intents such as '抓一下招投标' or '今日哪些能投', which increases the chance the skill activates when the user did not intend this specific workflow. Because the skill performs collection, analysis, and possible pushing of reports, accidental invocation can lead to unintended external access or dissemination.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill advertises automated collection from external platforms and report pushing, but it does not prominently warn users that it may contact external services and disseminate generated content. In context, this is more dangerous because the workflow targets multiple procurement platforms, uses caching, and includes scheduled pushing, creating a meaningful risk of unnoticed network activity and outbound sharing.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.