Openclaw Skill

Security checks across malware telemetry and agentic risk

Overview

The skill is coherent for production monitoring, but it exposes high-impact remediation, rollback, code indexing, and background auto-merge workflows that need careful review before installation.

Install only if you trust InariWatch with the connected production projects and code context. Use the least-privilege token available, prefer manual MCP configuration or inspect the npx setup package, verify exact alert and project IDs before approving fixes or rollbacks, and confirm whether auto-merge can be disabled or requires human review.

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: 90% confidence
Finding: The description "AI monitoring that fixes your code — query alerts, trigger remediations, rollback deploys, chat with your infrastructure" is broad enough to attract general requests about fixing code or infrastructure, not just scoped monitoring operations. Because this skill exposes high-impact capabilities such as triggering remediations and rolling back production, overly permissive invocation phrasing increases the chance of accidental routing into a skill that can perform sensitive actions.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal