ClawHub
Back to skill

Security audit

digital-legacy

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly local and purpose-aligned, but it handles very sensitive account and crypto information while overstating the strength of its encryption fallback.

Review before installing. If you use it, install the cryptography package before generating a will, test decryption, keep accounts.json private, avoid entering raw seed phrases or passwords into plaintext fields, and store the passphrase separately from the encrypted file.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
85% confidence
Finding
The skill documentation instructs users to run a script that reads and writes local files, but the skill declares no corresponding permissions. That mismatch can undermine user trust and permission enforcement because a host or reviewer may treat the skill as less capable than it really is, while the skill handles highly sensitive data such as account inventories, crypto wallet details, and encrypted will materials.

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The document asserts strong authenticated encryption with AES-256-GCM, but later states the script may fall back to a stdlib-only XOR-based stream cipher. That creates a dangerous mismatch between claimed and actual protection: users may believe highly sensitive digital-will data is securely encrypted when it may instead use a much weaker construction without equivalent confidentiality or integrity guarantees.

Intent-Code Divergence

Medium
Confidence
99% confidence
Finding
The code claims an AES-256-GCM 'stdlib fallback', but the fallback is actually a homegrown XOR-stream cipher with truncated HMAC. In a digital legacy tool handling highly sensitive account, wallet, and inheritance data, this is dangerous because users may believe they are getting standard authenticated encryption when they are instead relying on a nonstandard custom scheme that has not been peer reviewed and may fail under cryptographic scrutiny.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.