Open Wallet

WarnAudited by ClawScan on May 10, 2026.

Overview

This skill is openly designed for wallet transactions and signatures, but it gives agents broad raw wallet-request capability and can return signatures through redirects, so users should review each request carefully.

Install or use this only if you want agents to prepare crypto wallet requests for you. Before opening any generated link, independently verify the chain, wallet account, contract or recipient, transaction value, function/data, permissions, message text, and redirect destination. Never sign or approve a wallet prompt you do not understand.

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

An agent could craft a link for an irreversible transaction, token approval, batch call, or sensitive signature that the user might approve in their wallet.

Why it was flagged

The skill allows broad raw JSON-RPC wallet requests, including transaction and batch-call methods, rather than a narrowly scoped or allowlisted workflow.

Skill content
`method`: JSON-RPC method name ... `params`: URL-encoded JSON ... If `params` is a JSON array, it is treated as the exact JSON-RPC `params` array.
Recommendation

Only use links when you fully understand the chain, recipient/contract, function, value, permissions, and message being signed; do not approve unknown or unclear wallet prompts.

What this means

Approving a wallet prompt can spend funds, grant permissions, or create signatures that may be usable outside this chat.

Why it was flagged

The skill is explicitly designed to use the user's wallet authority. This is purpose-aligned and user-mediated, but it is still high-privilege account access.

Skill content
The page shows the request, prompts them to connect their wallet, switches to the requested chainId, then executes the JSON-RPC request.
Recommendation

Use a trusted wallet, verify the connected account and network, and reject any prompt that does not match the agent's plain-language explanation.

What this means

A signature or transaction result could be sent to an unintended website or prefilled message destination if the redirect URL is not trusted.

Why it was flagged

The result, including a signature, can be sent through an arbitrary redirect URL or deep link, but the instructions do not bound or require verification of the destination.

Skill content
If you include `redirect_url`, the app redirects after success or failure ... `result=<value>` ... `{{result_raw}}`: unencoded result string
Recommendation

Inspect redirect destinations before opening the link; prefer manual copy/paste for sensitive signatures unless the return URL is clearly trusted.

What this means

Users must trust the external wallet page to accurately display and execute the request.

Why it was flagged

The skill depends on an external hosted wallet UI, but the registry metadata does not provide source or homepage provenance for independent review.

Skill content
Source: unknown; Homepage: none
Recommendation

Use only if you trust tx.steer.fun and verify wallet prompts directly in your wallet before approval.