ClawHub
Back to skill

Security audit

I Have an Idea

Security checks across malware telemetry and agentic risk

Overview

This skill is a documentation workflow that persistently records ideas into Markdown files, with some chance of surprising users because it can trigger implicitly and create/update idea documents without asking each time.

Install this only if you want the agent to keep persistent Markdown idea records in your repository. Expect it to search existing Markdown docs and create or update an idea file when it interprets a message as a new idea; use explicit wording if you want immediate implementation instead of idea capture.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill explicitly instructs the agent to create or update Markdown files automatically as part of normal operation, while also forbidding asking the user whether to create the idea document. This creates an unauthorized file-write behavior risk: a user may intend brainstorming only, but the agent will persist content to the repository without an explicit, contemporaneous file-operation confirmation, which can surprise users and alter tracked project state.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The default prompt uses a very generic phrase ('capture my idea, develop it with me') combined with implicit invocation being enabled, which can cause the skill to trigger during ordinary brainstorming or casual discussion. That increases the chance of unintended routing and prompt-context takeover, where normal conversation is captured into the skill flow without clear user intent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.