ClawHub
Back to skill

Security audit

Comm LITE

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a read-only message-drafting helper, but its documentation also claims unrelated automation, file-writing, API, and network-troubleshooting capabilities that do not fit its declared purpose or permissions.

Install only if you intend to use it as a text-drafting aid. Do not treat its broader automation, API, networking, or file-writing claims as reviewed or supported, and avoid granting extra tools or credentials based on those sections.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

High
Confidence
96% confidence
Finding
The skill claims to be a pure Markdown, read-only communication helper with no external integrations, but elsewhere documents network/API behavior and command-style troubleshooting. This mismatch can mislead users and host agents about the trust boundary and expected behavior, increasing the chance that broader capabilities are granted or unsafe instructions are followed under false assumptions.

Description-Behavior Mismatch

High
Confidence
98% confidence
Finding
A skill presented as a simple message-drafting assistant later advertises generic automation, file parsing/writing, API integration, batch processing, and error recovery. This is dangerous because it expands the apparent operational scope far beyond the declared purpose, which can socially engineer operators into trusting the skill for actions or permissions unrelated to its stated function.

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The documentation instructs users to execute ping, check firewalls/proxies, and reason about network/API operations even though the skill is a communication drafting tool with only read access. Such unjustified operational guidance can prompt unsafe manual actions, normalize capability overreach, and blur the line between passive text assistance and infrastructure troubleshooting.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.