ClawHub Skill Guide — Scanner Compliance

• Outcome: Pass
• Latest audit: May 28, 2026, 1:50 AM
• Version:
  1.0.0

Overview

This is a documentation and template skill for publishing OpenClaw skills, with no evidence of hidden execution, credential access, persistence, or exfiltration.

Before using the publishing workflow, review the skill directory for secrets or unrelated files, and tighten any copied trigger descriptions so generated skills activate only for their intended tasks.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The guidance explicitly tells authors to include trigger keywords and activation scenarios, but it does not pair that advice with constraints on when the skill should not activate. In a skill whose purpose is helping content pass a security scanner, broad trigger phrasing can cause over-activation and make the skill influence unrelated tasks, including ones involving publishing, packaging, or security-sensitive configuration choices.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: Stating that the description is the primary trigger mechanism and recommending keywords matching user queries incentivizes maximizing activation breadth. Without equally strong guidance to limit scope, this can create ambiguous descriptions that cause the skill to be invoked in contexts beyond its intended domain, which is especially risky for a guide focused on passing security scans and publication workflows.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The template instructs authors to use generic placeholder trigger guidance like 'scenario1, scenario2, scenario3' rather than requiring precise invocation boundaries. In a skill ecosystem, vague descriptions can cause unintended or overly broad activation, which may lead users or orchestrators to invoke the skill in inappropriate contexts and increase the chance of unsafe downstream behavior.

Missing User Warnings

Low

• Confidence: 88% confidence
• Finding: The document provides `npx clawhub publish` commands that upload a local skill directory and associated metadata, but it does not explicitly warn users that publishing transmits local files to a remote service. In a skill-authoring context this can lead to accidental disclosure of secrets, test artifacts, or other unintended contents if users follow the workflow without first verifying what will be included.

VirusTotal

64/64 vendors flagged this skill as clean.