ClawHub
Back to skill

Security audit

Blog Writer

Security checks across malware telemetry and agentic risk

Overview

This blog-writing skill is coherent and disclosed, but users should understand it is designed to send drafts to a specific Notion database and retain finalized examples.

Install this only if you want a blog-writing workflow that can publish draft content to the configured Notion database. Do not use it with sensitive notes, confidential research, or unpublished proprietary material unless you are comfortable with that Notion publication step and with finalized posts being kept as future style examples.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
90% confidence
Finding
This utility includes an execution path that permanently deletes files from the skill's example library via `filepath.unlink()` when invoked with `prune --execute`. While the code appears intended for maintenance rather than abuse, destructive file-deletion capability is not necessary for a blog-writing skill's core function and creates avoidable risk if the script is run accidentally, pointed at unexpected content, or exposed through an agent workflow.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger conditions are broad enough that the skill could activate for vague writing requests and begin applying its workflow, including external publication steps, when the user did not explicitly ask for that behavior. In this context, unintended invocation matters because the skill is not purely local text generation; it can cause downstream side effects.

Missing User Warnings

High
Confidence
98% confidence
Finding
The skill mandates automatic publication of drafts to Notion and even labels failure to publish as workflow failure, but it does not require explicit informed consent at the time of transmission. This creates a real risk of exfiltrating sensitive drafts, research notes, or proprietary user content to an external service without a clear opt-in boundary.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.