Autonomous Wallet

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This is a high-impact crypto wallet skill that can use private keys and execute irreversible transactions, but its approval boundaries and external npm supply chain are not clearly bounded.

Review this carefully before installing. If you test it, use a new low-value wallet on a testnet first, pin and inspect the npm package, never provide your main recovery phrase, and require explicit confirmation for every transaction.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

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, installed package, or agent action is wrong or compromised, funds controlled by that key or recovery phrase could be transferred or approved irreversibly.

Why it was flagged

The skill asks for raw wallet credentials that can authorize blockchain transactions; combined with the wallet execution commands, this grants full control over the wallet's assets.

Skill content
PRIVATE_KEY ... Ethereum private key for wallet operations ... MNEMONIC ... 12/24-word recovery phrase
Recommendation

Do not use a main wallet seed. Use a dedicated low-value wallet or testnet, prefer hardware-wallet or multisig controls, and require explicit approval before every transaction.

What this means

A mistaken or manipulated intent could result in an unintended transfer, swap, staking action, or token approval.

Why it was flagged

The skill exposes natural-language commands that can move funds or grant token allowances, but the artifacts do not define mandatory human approval, recipient allowlists, spending caps by default, or other clear execution boundaries.

Skill content
autonomous-wallet execute "Send 1 ETH to alice.eth" ... autonomous-wallet execute "Approve Uniswap to spend 1000 USDC"
Recommendation

Only run transaction intents after reviewing the exact transaction details, enforce spending limits and simulation, and disable autonomous invocation for real funds unless strong guardrails are added.

What this means

The actual code that receives wallet credentials and signs transactions is outside the reviewed artifacts and could change in future npm releases.

Why it was flagged

The submitted skill contains no runnable code or install spec and relies on a globally installed external npm package with an unpinned version range to handle wallet operations.

Skill content
packages: ... name: openclaw-autonomous-wallet ... source: npm ... version: ">=0.1.0" ... npm install -g openclaw-autonomous-wallet
Recommendation

Pin an exact package version and integrity hash, review the package source before use, install in an isolated environment, and avoid exposing valuable keys to unreviewed code.

What this means

Users may be over-reassured into connecting valuable wallets or mainnet funds before independently verifying the software.

Why it was flagged

The skill makes broad security and privacy assurances, but the reviewed artifacts do not include the wallet implementation needed to verify those claims, and the README describes the project as alpha and not audited.

Skill content
This skill is **safe and verified** ... private keys never leave your device ... No data collection
Recommendation

Treat the safety claims as unverified until the exact code, audit status, package provenance, and key-handling behavior are independently confirmed.