Back to skill

Security audit

V19 Conformance Test Suite

Security checks for vulnerabilities and agentic risk

Overview

The skill is not clearly malicious, but it asks users to run an unbundled Python script that may contact an external service and create credentials without enough provenance or consent detail.

Review this before installing or running it. Do not run V19_Conformance_Test_Suite.py unless you have the actual script from a trusted source and understand which service it contacts, what data it sends, and what credentials it creates or stores.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

The natural-language content that describes the skill's purpose and execution steps is presented only in Chinese, while also claiming applicability to "any external developer." For a broadly targeted skill, forcing a single language without offering alternatives or an opt-in can violate language/locale policy requirements.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly tells users to run a zero-configuration script and states that it includes a public V19 experience key, but it does not clearly warn that execution will contact an external Cloudflare-hosted service and perform self-registration to obtain credentials. This creates informed-consent and supply-chain risk: users may expose their environment to unreviewed outbound network activity and unintentionally mint or handle keys without understanding the consequences.

Content

No source excerpt is available for this finding.