Lynkr AI Routing Proxy

• Outcome: Review
• Latest audit: May 29, 2026, 9:27 PM
• Version:
  0.6.0

Overview

Lynkr appears to be a legitimate LLM routing proxy, but it routes broad OpenClaw model traffic through a separately installed proxy with incomplete privacy, persistence, and tool-execution boundaries.

Review before installing. Use Lynkr only if you are comfortable routing OpenClaw model traffic through it and its configured upstream providers. Pin and verify the npm package, keep the proxy bound to localhost, use least-privilege provider credentials with spending limits, and review or disable server-side tool execution and persistent session features until their storage and approval behavior is clear.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: This skill is an LLM routing proxy that forwards user prompts and associated routing metadata to multiple external model providers, but the description does not clearly warn users that their requests may leave the local environment and be processed by third parties. In a coding/agent setting, prompts can contain source code, secrets, internal architecture details, or sensitive workflow metadata, so lack of disclosure can lead to unintended data exposure and policy noncompliance.

VirusTotal

66/66 vendors flagged this skill as clean.