ClawHub
Back to skill

Security audit

天气

Security checks across malware telemetry and agentic risk

Overview

The skill presents itself as a weather helper, but it requests broad read, exec, and write authority while also giving unrelated automation and workflow guidance.

Review this skill carefully before installing. It does not show malicious behavior, but it asks for broad system tools and contains generic automation instructions that do not fit a weather helper. Use only for explicit weather queries and avoid letting it run commands or modify files unless the action is clearly necessary and user-directed.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

High
Confidence
83% confidence
Finding
The skill uses overly broad and inconsistent activation guidance that extends far beyond weather, which can cause an agent to invoke it in unrelated contexts. In an agent environment with read/exec/write tools enabled, ambiguous routing increases the chance of unnecessary system actions, misuse of credentials, or unsafe execution under the wrong user intent.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The open-ended trigger phrase invites invocation whenever the user wants to improve efficiency or automate workflows, without limiting the skill to weather operations. Because the skill advertises powerful tools like exec and write, this broad trigger surface can steer the agent into system-impacting behavior unrelated to the user's actual need.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The markdown references command execution, file operations, and API credential setup while exposing tools read, exec, and write, but it does not provide strong, explicit guardrails about system modification, command safety, or secret handling. In a tool-using agent, this can normalize risky behavior and make accidental destructive commands, unsafe file writes, or credential leakage more likely.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.