ClawHub
Back to skill

Security audit

Charge My Users

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed SettleMesh billing-integration guide with payment-sensitive login behavior that users should understand before use.

Install this only if you intend to use SettleMesh for end-user billing. Before first use, confirm the agent is logging into the intended SettleMesh account, review any charge or allowance prompts carefully, and never treat API keys as payer tokens.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The skill explicitly tells the agent to run `settlemesh login` on its own, open a browser, poll for approval, and cache the resulting session if `SETTLE_API_KEY` is unset. That expands the skill from billing guidance into credential acquisition and session persistence, which can cause the agent to initiate authentication flows and store credentials without a clear, contemporaneous user approval step.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill directs the agent to initiate an interactive login flow automatically and specifically says not to stop to ask the user to log in manually. Even if browser approval is required, suppressing an explicit user-facing warning at the moment authentication begins undermines informed consent and can lead to unexpected credentialed sessions being established.

Static analysis

No suspicious patterns detected.