query-token-audit Skill (Binance) for Claude: What It Does and How to Install It

Makes Claude check a token contract for scams, honeypots and odd taxes using Binance Web3's audit endpoint.

Installs: 10.8K; on skills.sh
Stars: 1.1K; on GitHub
Updated: Oct 2026; last repo commit

This skill sends one token contract address to a Binance Web3 audit endpoint and reports what comes back: a risk level, buy and sell tax, whether the contract code is verified, and the list of checks that hit. It supports BSC, Base, Solana and Ethereum, and it loads when you ask whether a token is safe or before a swap.

SKILL.md's request has no API key or auth header. It runs a curl command with a fresh UUID request ID, so it needs a shell with curl and uuidgen. A LOW result is not a safety guarantee, and SKILL.md tells Claude to give risk flags only and add a not-investment-advice line.

Good for

Crypto traders checking a token before a swap
People screening a contract address on BSC, Base, Solana or Ethereum
Developers adding a pre-trade safety step to an agent

Install query-token-audit

Claude Code

Copy this prompt.

Install the query-token-audit skill in Claude Code. Run this command:

npx skills add https://github.com/binance/binance-skills-hub --skill query-token-audit

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

Pick one route. The skills CLI route installs just this skill; the README's own command adds the whole Binance Skills Hub. The README lists Node.js 22 or higher as the installer prerequisite. The skill needs a shell, so the box has no Claude app upload.

Built from the package data we hold. Not yet tested by us.

What's in the skill

When it loads

Query token security audit to detect scams, honeypots, and malicious contracts before trading. Returns comprehensive security analysis including contract risks, trading risks, and scam detection. Use when users ask "is this token safe?", "check token security", "audit token", or before any swap.

Sections in SKILL.md

Overview
Use Cases
Supported Chains
API: Token Security Audit
Method: POST
User Agent Header
Notes

Key facts

Author: binance-web3-team (SKILL.md metadata). It sits in Binance's own binance/binance-skills-hub repo
License: No license stated in SKILL.md or the README
Installs on skills.sh: 10,846 (Nov 2, 2026)
GitHub stars: 1,115 for the whole binance/binance-skills-hub repo (Nov 2, 2026)
Last repo commit: Oct 26, 2026
What it includes: 1 file: SKILL.md only (no scripts)
Needs: A shell with curl and uuidgen. The request has no API key
Skill version: 1.4 (SKILL.md metadata)
Works in: Agents such as OpenClaw and Claude Code, per the README

What it includes

POST one contract address and read the risk report
Sends binanceChainId, contractAddress and a UUID requestId to web3.binance.com with a curl command.
Try: “Is 0x55d398326f99059ff775485246999027b3197955 safe on BSC?”
Runs code · Token Security Audit API
BSC (56), Base (8453), Solana (CT_501), Ethereum (1)
The chain ID is part of the request.
Read · Supported chains
Maps riskLevel 0 to 5 to LOW, MEDIUM or HIGH and an action
Level 4 is high risk, avoid trading. Level 5 is severe, block the transaction. LOW is not a guarantee of safety.
Read · Risk level reference
Only show results when data is available and supported
If hasResult or isSupported is false, Claude says no audit data is available and shows no risk fields.
Read · Response handling
Over 10% is critical, 5 to 10% a warning, under 5% acceptable
Applied to the buy and sell tax in the result.
Read · Tax thresholds
Risk flags only, no investment advice, closing disclaimer
SKILL.md tells Claude to always end with a disclaimer that the result is for reference only.
Read · Output rules

From the official docs

Permissions

Low risk

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

The address you check is sent to Binance
The contract address, chain ID and a fresh UUID go to the web3.binance.com audit endpoint, with a User-Agent of binance-web3/1.4 (Skill). Do not check addresses you want to keep private. github.com
A LOW result is not a safety guarantee
SKILL.md says results are point-in-time snapshots and that teams can change contracts or restrict liquidity after purchase. The README disclaimer says outputs are informational, not advice. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	No read-only switch. SKILL.md has one audit request.
Talks to	web3.binance.com
Sign-in and keys	SKILL.md's example request has no key, token or login header.
Scopes	Claude runs a curl POST to web3.binance.com with the chain ID, contract address and a request ID Results come back as a risk level, tax figures and risk check hits

Questions

What does the query-token-audit skill do?

It checks one token contract against Binance Web3's security audit and reports the risk level, taxes, code verification and which risk checks hit. Claude uses it before you buy or swap.

How do I install query-token-audit?

Use the install box at the top of this page. It installs just this skill, or the whole Binance Skills Hub if you want it. The skill needs a shell for curl, so there is no Claude app upload.

Does it need an API key or a Binance account?

SKILL.md's example request carries no key, token or login, only content-type, source, accept-encoding and user-agent headers. The README says certain other Binance skills' endpoints need API credentials, which SKILL.md for this skill does not mention.

Does a LOW risk result mean the token is safe?

No. SKILL.md says "LOW risk does NOT mean "safe." Audit results are point-in-time snapshots." Project teams can change contracts or restrict liquidity after you buy.

Which chains does it cover?

BSC, Base, Solana and Ethereum. For a token with no audit data, SKILL.md has Claude say so instead of showing a risk level.

How is it different from crypto-market-rank?

This skill audits one contract. crypto-market-rank, a sibling in the same repo, returns ranked leaderboards such as trending tokens and smart-money inflow through a Node script. Neither SKILL.md requires the other.

Is it safe?

It sends the contract address and chain ID you give it to Binance's web3.binance.com endpoint with a curl command. SKILL.md says to output risk flags only, no investment advice. The README disclaimer says outputs are informational and not financial advice.

Information

updated Nov 2, 2026

Repository: binance/binance-skills-hub
Folder: skills/binance-web3/query-token-audit

binance/binance-skills-hub skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 2

query-token-audit SKILL.md · github.com
binance-skills-hub README · github.com