clawcash

Security checks across malware telemetry and agentic risk

Overview

The skill is open about being a credit/payment tool, but its install flow asks an agent to perform identity onboarding, token minting, package execution, and financial setup without step-by-step user approval.

Review this carefully before installing. Only proceed if you intend to create or use ClawLens/ClawCash financial access, complete any required identity checks, and allow a CLI to store tokens locally. Require the agent to pause before package execution, token minting, credit activation, paid x402 calls, or repayment, and treat platform tokens and repayment URLs as secrets.

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)

Vague Triggers

Medium
Confidence
96% confidence
Finding
The trigger phrase 'install ClawCash (or equivalent)' is overly broad and can cause unintended activation of a workflow that performs package execution, identity onboarding, token minting, and external network calls. In an agent setting, ambiguous activation language increases the risk that a casual or indirect user utterance is interpreted as authorization for sensitive actions.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill instructs the agent to autonomously install dependencies, execute third-party code via npx, perform KYC-related setup, and mint access tokens 'in one turn' without upfront warning or stepwise approval. This removes meaningful user consent around credential creation and external transmissions, making accidental secret exposure or unwanted account actions more likely.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal