OpenClaw Risk Manager

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only finance risk-management skill with broad activation wording, but it does not contain code, credentials, broker access, or hidden data access.

Install this only for analysis, checklists, calculators, and risk-management guidance. Do not let it place, modify, or cancel trades through any separate brokerage or automation tool unless you explicitly request and confirm that action.

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 (2)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The metadata explicitly says to use the skill proactively for broad categories like risk assessment, trade tracking, or portfolio protection, which can cause the agent to invoke it in situations where the user did not request specialized portfolio-risk handling. Over-broad activation increases the chance of inappropriate routing, unnecessary tool/skill use, and finance-specific guidance being applied to adjacent tasks without sufficient user intent or context.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The usage guidance is generic ('working on risk manager tasks or workflows' and 'needing guidance, best practices, or checklists'), so the skill may match a wide range of loosely related requests. In a financial context, unintended invocation is more dangerous because the model may provide authoritative-seeming risk or trading guidance in cases where the user asked something only tangentially related.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal