ClawHub
Back to skill

Security audit

Contract Agent Free

Security checks across malware telemetry and agentic risk

Overview

The skill is a contract automation tool, but it needs Review because it mixes simulated escrow with real-payment language, uses broad activation wording, and handles private signing keys with write/exec authority.

Install only if you understand this is not a real escrow/payment system in the free edition. Treat contract signing, simulated funds, and private keys as sensitive: restrict use to explicit contract tasks, verify any external package before running npm install, avoid real-money reliance, and store keys in an encrypted or managed keystore rather than plain local files.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The document describes escrow-style fund locking and automatic release as if they are real financial controls, but elsewhere states the free edition only supports a simulated escrow account. This mismatch can mislead users or calling agents into treating non-binding simulation output as real custody or payment execution, creating transaction, compliance, and trust failures.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger condition is so broad ('use when AI model calling, intelligent dialogue, agent orchestration, LLM applications are needed') that the skill may be invoked outside its narrowly appropriate contract-management context. In an agentic environment with write and exec tools enabled, over-broad activation increases the chance of unintended execution paths, incorrect contract actions, or unsafe automation around sensitive business workflows.

Missing User Warnings

High
Confidence
96% confidence
Finding
The skill describes behaviors like fund deposit locking, irreversible withdrawal prevention, and automatic release without a prominent warning about financial consequences and edition limitations. In a contract/payment context, users or other agents may rely on these statements to make real business decisions, causing financial loss or unauthorized operational commitments if the workflow is misunderstood or auto-executed.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill instructs generation of RSA private keys and notes they must be stored securely, but the operational guidance is not prominent or restrictive enough given that local storage and export are part of the workflow. In an agent environment with read/write/exec capabilities, weak handling of private keys can lead to credential theft, signature forgery, and impersonation of agents in contractual actions.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.