Back to skill

Security audit

ClawWeather Pro

Security checks for vulnerabilities and agentic risk

Overview

The skill is mainly a weather helper, but it bundles an under-disclosed paid trading-weather client that can use a spending-capable API key.

Review before installing. The ordinary weather features send city or hub data to third-party weather services, which is expected. Do not set X402_API_KEY or X402_BASE unless you intentionally want the paid premium trading-weather integration, understand where the key will be sent, and are prepared for per-call charges. The publisher should align the manifest and documentation with the bundled premium client or remove it.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

The skill is presented as a free, no-API-key weather tool, but the documentation also references a paid call, a spending-capable key, and a configurable external backend for premium trading-weather data. That mismatch is dangerous because users and reviewers may grant trust or permissions based on the 'free/no key' description while the skill can trigger billable requests or exfiltrate a sensitive API key to a non-declared endpoint.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The documentation claims a paid weather call exists even though the manifest only describes free wttr.in/Open-Meteo usage with no API key or paid calls. In practice, this creates security and trust issues: operators may not provision controls for billing, secret handling, or outbound access because the manifest understates the skill's capabilities.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

This file implements a paid x402 'trading weather' client that conflicts with the skill metadata, which describes a free general weather service using wttr.in and Open-Meteo with no API key. That mismatch is dangerous because users may unknowingly trigger paid financial-data-oriented network calls and disclose a spending-capable API key to an unrelated third-party service outside the declared trust boundary.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The code adds finance/trading-oriented capability that is not justified by the stated purpose of ordinary weather and forecast retrieval. In this context, hidden or unnecessary expansion into market-related functionality increases user surprise, broadens data-sharing and billing risk, and weakens trust in the skill's declared behavior.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The script makes an outbound network request to a third-party service, which creates a data-sharing and availability dependency outside the local trust boundary. Although this file sends only fixed coordinates for predefined hubs rather than arbitrary user-entered city names, it still exposes usage patterns and relies on an external endpoint that could log requests or fail unexpectedly.

Content

Scanner excerpt · scripts/trading_weather.py (reported line 22)May include surrounding context.

python
def get_open_meteo(lat: float, lon: float) -> dict:
    url = (f"https://api.open-meteo.com/v1/forecast?latitude={lat}&longitude={lon}"
           "&current_weather=true&daily=weathercode,temperature_2m_max,temperature_2m_min&timezone=auto")
    try:
        with urllib.request.urlopen(url, timeout=15) as r: