Agent Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed blockchain game, but it asks an agent to control a real crypto wallet, grant very broad USDC approval, and potentially run an unpinned CLI automatically.

Only install this if you understand it gives the agent practical control over a crypto wallet. Use a fresh, low-balance wallet, pin and review the CLI, avoid unlimited approvals where possible, and do not enable auto/cron mode unless you set clear spending limits and stop conditions.

VirusTotal

66/66 vendors flagged this skill as clean.

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.

#
ASI03: Identity and Privilege Abuse
High
What this means

If the private key or wallet authority is mishandled, the user can lose real cryptocurrency funds.

Why it was flagged

The skill explicitly uses a private key for a real crypto wallet, which gives the agent authority to spend funds and sign blockchain transactions.

Skill content
**This skill manages a self-custodial wallet with real funds on Base.** ... Store `BASE_PRIVATE_KEY` only in environment variables or secured config files
Recommendation

Use only a dedicated low-balance wallet, never a main wallet, and verify that credential requirements and private-key handling are acceptable before installing.

#
ASI02: Tool Misuse and Exploitation
High
What this means

The agent may authorize broad USDC spending, not just one heartbeat or registration payment.

Why it was flagged

Automatic unlimited token approval is much broader than a single payment and can expose all approved USDC to contract or CLI misuse.

Skill content
The CLI automatically checks USDC allowance before `register` and `heartbeat` commands. If insufficient, it approves `maxUint256` before proceeding.
Recommendation

Prefer capped allowances, review the contract and CLI code, and set a small dedicated wallet balance before allowing these commands.

#
ASI04: Agentic Supply Chain Vulnerabilities
Medium
What this means

A future or compromised CLI release could handle the wallet private key and transactions differently from what the skill describes.

Why it was flagged

The skill permits running an unpinned @latest CLI package through Bash, while the provided artifact set contains no reviewed CLI code.

Skill content
allowed-tools: ["Bash(npx last-ai-standing-cli@latest *)", "Bash(las *)"]
Recommendation

Pin the CLI to a reviewed version and inspect the package source before using it with a funded wallet.

#
ASI10: Rogue Agents
Medium
What this means

The agent could continue spending USDC and gas over time to stay alive in the game.

Why it was flagged

The skill encourages a recurring autonomous mode for ongoing gameplay, but the visible instructions do not define budget limits, user confirmations, or a stopping condition.

Skill content
# Or use auto mode (recommended for cron)
las auto
Recommendation

Run auto mode only with explicit user approval, a limited wallet balance, and clear stop conditions.