ClawHub
Back to skill

Security audit

Process Output

Security checks across malware telemetry and agentic risk

Overview

This skill only changes assistant replies into structured progress JSON blocks, which is disclosed and not harmful by itself but may be incompatible with normal chat use.

Install this only if you want the assistant to wrap most replies in machine-parseable progress JSON for a compatible client. Avoid enabling it in normal chat contexts unless users know how to opt out with a final-answer-only request.

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

High
Confidence
96% confidence
Finding
The skill declares itself as the default output mode for essentially any user message, which overrides normal conversational behavior and forces unsolicited protocol emissions. This broad activation can disrupt downstream consumers, leak internal reasoning/progress metadata into contexts that did not request it, and create prompt-injection leverage by making the model follow skill formatting over user intent.

Natural-Language Policy Violations

Medium
Confidence
94% confidence
Finding
The skill forces a custom `openclaw-process` protocol in all replies without user opt-in, compelling the assistant to emit structured intermediate process data even for simple interactions. This is dangerous because it can violate user expectations, break clients that expect plain text, and encourage disclosure of internal workflow/progress details that should remain minimal unless explicitly requested.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.