Openclaw Model Orchestrator

• Outcome: Pass
• Latest audit: May 28, 2026 at 1:04 PM UTC
• Version:
  0.1.1

Overview

This skill appears to do what it advertises, but users should remember that orchestration sends task content and intermediate outputs to multiple configured LLM providers.

Install only if you are comfortable with the task text, code snippets, and intermediate outputs being routed to the model providers configured in OpenClaw. Avoid using it for secrets, regulated data, or private customer material unless those providers are approved for that data.

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: 94% confidence
• Finding: The orchestrator forwards user tasks, intermediate outputs, and reviewer synthesis content to multiple external models through `api.inference`, potentially multiplying disclosure across providers. Because the code does not present a clear user-facing warning, consent gate, or provider/data-boundary notice before sending content, users may unknowingly expose sensitive prompts, code, secrets, or internal data to third-party services.

VirusTotal

No VirusTotal findings