Back to skill

Security audit

Universal Profile

Security checks across malware telemetry and agentic risk

Overview

This looks like a legitimate Universal Profile tool, but it can use local controller keys to submit irreversible blockchain actions with limited built-in safeguards.

Install only if you are comfortable letting this skill access a Universal Profile controller key. Use a testnet or low-value, least-privilege controller first; avoid full-access permissions; keep key files locked down; and manually review every transfer, mint, relay, batch, permission, and authorization action before allowing an agent to run it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (11)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill exposes operational capabilities involving environment variables, network access, and shell-oriented workflows, but it does not declare explicit permissions or boundaries. That mismatch increases the chance an agent can access credentials, invoke external services, or perform local actions without clear user consent or sandbox policy, which is risky in a blockchain wallet-management skill.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The skill description is very broad and covers identity, permissions, tokens, and blockchain operations across multiple chains without clear trigger phrases or activation limits. In an agent setting, vague scope can cause over-invocation and unsafe execution of sensitive actions like signing, transferring value, or changing permissions when the user's request is ambiguous.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documentation instructs users to store a controller private key in a local JSON config file, which materially increases key-exposure risk through file leakage, backups, logs, malware, or accidental disclosure. Because this skill manages on-chain identity and permissions, compromise of that key can directly enable unauthorized transactions, permission changes, or loss of assets.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The examples demonstrate high-impact operations such as granting permissions, revoking permissions, and transferring value through a Universal Profile without clearly warning users that these actions can change account control or move funds irreversibly. In an identity and blockchain operations skill, omission of safety guidance materially increases the chance of accidental misuse, especially by users copying example code directly into production or test environments with real assets.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
Referencing a controller private key via an environment variable is a common pattern, but presenting it without any warning about credential sensitivity normalizes direct handling of highly privileged secrets. In the context of Universal Profiles, compromise of the controller key can lead to full profile takeover, permission changes, and unauthorized asset movement across supported chains.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The examples show live deployment, minting, and transfer operations that create irreversible on-chain state changes, but they do not warn users about gas costs, permanent asset movement, or the risk of using production addresses and wallets. In a skill intended to manage blockchain identities and tokens, omission of transaction-safety warnings increases the likelihood of accidental loss, unintended minting, or deployment to the wrong network/account.

Missing User Warnings

High
Confidence
97% confidence
Finding
The example explicitly uses `force: true` and labels it as bypassing the LSP1 Universal Receiver check, which suppresses a compatibility/safety mechanism intended to protect recipients from unsupported token receipts. Without a strong warning, users may transfer or mint assets to addresses that cannot properly handle them, causing asset loss, stuck tokens, or broken application flows.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The authorization flow generates a URL intended to add the tool as a controller and only provides generic approval instructions, without a prominent warning that controller permissions can grant powerful, potentially wallet-draining or profile-modifying access. In the context of a blockchain identity/permissions skill, insufficiently explicit consent UX is dangerous because users may authorize high-impact permissions without understanding that they are delegating control over their Universal Profile.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The helper returns the controller private key as part of a broad convenience object, which unnecessarily expands secret exposure to every caller of this function. In an agent skill that manages blockchain identities and signing across multiple chains, any downstream logging, serialization, debugging, prompt injection, or tool misuse can leak the key and enable full account compromise and irreversible on-chain transactions.

Unpinned Dependencies

Low
Category
Supply Chain
Content
"author": "frozeman",
  "license": "MIT",
  "dependencies": {
    "ethers": "^6.9.0",
    "viem": "^2.45.1"
  },
  "devDependencies": {
Confidence
91% confidence
Finding
"ethers": "^6.9.0"

Unpinned Dependencies

Low
Category
Supply Chain
Content
"license": "MIT",
  "dependencies": {
    "ethers": "^6.9.0",
    "viem": "^2.45.1"
  },
  "devDependencies": {
    "eslint": "^8.56.0",
Confidence
91% confidence
Finding
"viem": "^2.45.1"

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.