Back to skill

Security audit

Global Tax Guide

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only tax guidance skill with no hidden execution or persistence, though users should verify its tax guidance before acting.

Install only if you want educational ecommerce tax triage. Treat its thresholds, rates, marketplace facilitator coverage, and registration links as starting points, not professional advice; verify current rules with official tax authorities or a qualified tax professional before registering, filing, collecting tax, or changing business operations.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

This skill provides detailed jurisdiction-specific tax compliance guidance and only introduces the 'educational, verify professionally' disclaimer late in the workflow. Users may reasonably rely on the output as actionable legal or tax advice before reaching that caveat, increasing the chance of noncompliant registrations, filings, or tax treatment based on incomplete or outdated rules.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

md
|---|---|---|---|
| Jurisdiction to assess first | Highest revenue market above threshold | Market approaching threshold | Market with negligible volume |
| Threshold status signal | Revenue + transaction count both clear | Revenue clear, count borderline | Only one metric tracked |
| Marketplace facilitator reliance | Platform confirmed MF in writing | Platform docs say MF applies | Assumed MF without confirmation |
| Filing scheme choice | OSS/IOSS for EU multi-country | Individual country VAT registrations | No scheme, selling anyway |
| Product classification approach | Category confirmed with local customs code | Same category used across all regions | Default to standard rate everywhere |
| Professional support trigger | Revenue exceeds $50k in new market | Launching into regulated product category | Any cross-border expansion |

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

This markdown file provides direct registration links and threshold guidance that could materially affect tax registration and filing decisions. Although L003 notes that thresholds change, the document does not include a stronger user-facing warning near the actionable registration section that this is not legal or tax advice and should be verified with current official sources or a qualified professional before taking action.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.