Agent Forge

Security checks across malware telemetry and agentic risk

Overview

This skill is a legitimate agent-creation helper, but it can make persistent agent and gateway permission changes after broad triggers without a clear final approval gate.

Install only if you intentionally want a skill that can create persistent OpenClaw agents and modify multi-agent communication settings. Before using it, review the exact files and gateway patch, keep session visibility and tools as narrow as practical, check what is in USER.md before it is copied, and use simple trusted agent IDs.

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: 94% confidence
Finding: The trigger phrases are broad enough to match ordinary user requests such as '创建一个agent', which can cause the skill to activate unexpectedly. In this skill, unintended activation is more dangerous than usual because activation leads to agent creation, file generation, configuration changes, and potential permission assignment.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal