zscore

PassAudited by ClawScan on May 10, 2026.

Overview

This appears to be a legitimate Zeru registry skill, but it can use a wallet private key to spend gas/fees and publish or update on-chain agent data.

Install only if you intend to manage Zeru registry entries. Use a dedicated low-balance wallet, test on Base Sepolia first, review fees and JSON before write commands, and treat remote agent metadata as untrusted.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If the skill is invoked for writes, it can sign Zeru API messages and send blockchain transactions from the configured wallet, spending gas and possibly registration fees.

Why it was flagged

The CLI reads a wallet private key from the environment and creates an ethers signer for write operations.

Skill content
const pk = process.env.PRIVATE_KEY; ... const signer = new Wallet(pk, provider);
Recommendation

Use a dedicated low-balance wallet, avoid using a primary wallet key, and prefer Base Sepolia for testing.

What this means

A mistaken or unintended registration can spend ETH for the fee and gas and create persistent on-chain/public agent records.

Why it was flagged

Registration sends a payable on-chain transaction to the configured identity registry.

Skill content
const tx = await registry.registerWithFee(agentURI, { value: fee });
Recommendation

Review the network, fee, wallet, and agent JSON before running register, set-metadata, or unset-wallet.

What this means

Reading unknown agents may contact arbitrary external URLs and place untrusted metadata into the agent's output context.

Why it was flagged

Reading an agent can trigger an HTTP fetch to the agentURI stored on-chain, which may be controlled by another registrant.

Skill content
if (agentURI && agentURI.startsWith("http")) { const res = await fetch(agentURI); ... result.parsedJson = (await res.json()); }
Recommendation

Treat fetched agent metadata as untrusted and consider adding URL allowlisting or user confirmation before fetching arbitrary agentURI values.

What this means

Installing npm dependencies can introduce normal package supply-chain risk.

Why it was flagged

The skill relies on npm package installation before use. This is expected for the Node-based CLI, and package versions are pinned, but it is still an external dependency step.

Skill content
Run once to install dependencies:

cd {baseDir} && npm install
Recommendation

Install from a trusted environment, review package.json, and consider using a lockfile or reproducible install process.