Back to skill

Security audit

okkmax-find-relay

Security checks for vulnerabilities and agentic risk

Overview

This skill performs clearly scoped relay comparisons without requesting API keys, changing local settings, or running code.

Expect model searches and comparison criteria to reach OkkMax. Treat its measurements as dated third-party evidence, not guarantees about a relay's safety. The skill needs no API keys. Tool text may appear in English for languages other than Chinese.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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 (1)

The workflow forces English locale for every non-Chinese user without offering a locale choice.

Low
Category
Output Handling
Confidence
86% confidence
Finding

The instruction requires locale "en" on every tool call for all users who are not Chinese, including users communicating in other languages. Although line 8 requires replies in the user's language, this separate tool-locale constraint has no user opt-in or documented justification.

Content

No source excerpt is available for this finding.