openclaw configurator

Security checks across malware telemetry and agentic risk

Overview

The available evidence points to a configuration-writing helper with a weak evaluation prompt, not hidden or malicious behavior.

Before installing, treat generated configuration as a draft: make sure the skill asks for your intended use, identity/personality preferences, safety boundaries, and target file location before writing or replacing any files.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The prompt "给我写个配置。" is so underspecified that it encourages the skill to guess user preferences and generate a default configuration without collecting key safety and identity requirements. In a configuration-generating skill, this can produce unsafe or misaligned agent files, especially when the expected behavior is supposed to require multi-round clarification before producing outputs.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal