Security audit

Alephnet Node Manager

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly coherent, but it gives agents command-line access to shared memory and wallet operations without enough confirmation and privacy guidance.

Review this skill before installing. Only use it with a trusted alephnet-node CLI and trusted npm packages, keep API keys out of chat and logs, prefer user or conversation memory scopes unless sharing is intended, and require manual confirmation before any wallet.send, wallet.stake, memory.sync, global query, or agent/team activation command.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Intent-Code Divergence

Medium

Confidence: 90% confidence
Finding: The documented output schema describes a grading/audit report rather than the results of node-management, memory, agent, wallet, or network operations described elsewhere in the skill. This mismatch can cause downstream agents or users to mis-handle operational results, ignore failures, or make unsafe assumptions about whether sensitive actions like token transfers or memory synchronization actually succeeded.

Intent-Code Divergence

Medium

Confidence: 89% confidence
Finding: The input format section defines a generic content/strict_level review interface that contradicts the rest of the skill, which documents command-style operational actions over memory, agents, teams, and wallets. Such interface confusion is security-relevant because an integrating agent may route arbitrary content into the skill or invoke the wrong workflow, leading to unintended execution of state-changing operations or bypass of expected validation.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill instructs users to place API keys in config files and environment variables and also describes external LLM, network, and storage integrations, but it does not warn that these secrets may be transmitted, logged, or exposed through command execution. In a skill with exec capability and broad network interactions, missing secrecy guidance materially increases the risk of credential leakage and unauthorized access to third-party services.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The documented memory and synchronization operations affect visibility, persistence, and sharing scope of potentially sensitive data, yet the skill does not warn users about privacy, access scope, or rollback implications. In particular, creating shared memory fields, global queries, and syncing conversation context can expose private or organizational information beyond the user's intent.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: Wallet staking and token transfer commands have direct financial consequences, but the skill omits warnings about irreversibility, lock periods, recipient verification, fees, and loss risk. In an exec-enabled operational skill, users or calling agents may execute these commands without adequate confirmation, resulting in unintended asset movement or funds being locked for extended periods.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.