ClawHub
Back to skill

Security audit

Ziptax Tool Free

Security checks across malware telemetry and agentic risk

Overview

This sales-tax lookup skill is not clearly malicious, but its broad routing, write/exec authority, unrelated operation claims, and persistent API-key guidance need review before use.

Install only if you intend to use zip-tax.com for U.S. sales-tax lookups. Limit use to address, ZIP, coordinate, and metrics queries; do not rely on the unrelated create/modify/delete/import/export language. Store the API key using your normal secret-management practice instead of casually appending it to a shell profile, and avoid submitting sensitive personal addresses unless third-party transmission is acceptable for your use case.

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 Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The skill presents itself as a narrowly scoped sales-tax lookup tool, but the documentation adds generic create/modify/delete/export/reset/import operation modes that are unrelated to the declared functionality. This kind of scope inflation can cause an agent or user to over-trust the skill and invoke it in unintended contexts, increasing the chance of unsafe or unauthorized actions.

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The documentation claims support for create/query/modify/delete-style operations, while the included implementation only shows read-only lookups and metrics retrieval. This mismatch is dangerous because agent systems may rely on the higher-level documentation when choosing or granting capabilities, leading to overbroad invocation or trust in nonexistent controls.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The manifest includes a very broad invocation phrase tied to generic coding, debugging, and deployment help requests, which goes far beyond sales-tax lookup. Overbroad routing language can cause the skill to be selected in unrelated developer workflows, exposing API keys, user data, or command execution paths where this tool does not belong.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The capability coverage section lists excessively broad keywords, including generic development-assistance terms, which can cause unintended auto-invocation. In an agent ecosystem, this increases the attack surface by allowing the skill to activate outside its legitimate domain and influence workflows unnecessarily.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs users to send addresses, ZIP codes, coordinates, and an API key to an external service, but does not prominently warn about third-party transmission, data sensitivity, retention, or privacy implications. This is risky because location data and credentials may be disclosed to an external provider without informed consent or organizational approval.

External Transmission

Medium
Category
Data Exfiltration
Content
exit 1
fi
# ...
API_URL="https://api.zip-tax.com/request/v60"
# ...
# 解析参数
while [[ $# -gt 0 ]]; do
Confidence
93% confidence
Finding
https://api.zip-tax.com/

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.