ClawHub
Back to skill

Security audit

Lynkr AI Routing Proxy

Security checks for vulnerabilities and agentic risk

Overview

Lynkr is a disclosed LLM routing proxy, but users should understand that coding requests may be routed to cloud providers and logged locally.

Install only if you are comfortable routing coding prompts, tool outputs, and possibly source snippets through the providers you configure. Prefer local-only providers for sensitive work, review the npm package and runtime settings, and check how to disable or limit telemetry and memory before using it with private code.

SkillSpector

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)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill advertises SQLite-backed routing telemetry, raw telemetry endpoints, latency tracking, and quality scoring, but it does not prominently warn users that request-related metadata may be collected and stored. In an LLM proxy handling coding traffic, that metadata can reveal sensitive prompts, repository characteristics, provider usage, and operational behavior, creating privacy and confidentiality risks if users deploy it without informed consent.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill is a universal proxy that can route requests to many external cloud providers, yet the description does not clearly warn that prompts, code, and other request contents may be transmitted to third-party services. In a coding-assistant context, this is especially sensitive because users may unknowingly send proprietary source code, secrets, or regulated data off-host based on automatic routing decisions.

Static analysis

No suspicious patterns detected.