local-config-model-recommender

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:19 PM UTC
• Version:
  1.0.0

Overview

This skill is a simple model-recommendation guide that reads local OpenCLAW model configuration, with no evidence of hidden execution, persistence, or data exfiltration.

Install only if you are comfortable with the agent reading your local OpenCLAW configuration to identify available model IDs. Avoid asking it to print or share the full configuration, especially if that file contains endpoints, provider details, tokens, or other sensitive settings.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• 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

Low

• Confidence: 91% confidence
• Finding: The skill explicitly states it will read the user's local OpenCLAW configuration, which is local data access, but it does not clearly warn about that access, what file is read, or what information may be exposed through recommendations. This creates a transparency and consent issue: users may invoke the skill without realizing it inspects local configuration that can reveal provider choices, model inventory, and potentially sensitive configuration structure.

VirusTotal

66/66 vendors flagged this skill as clean.