Which AI agents can use CryptoGuard?

CryptoGuard is designed as a pre-transaction hook for AI coding agents like Claude Code and OpenAI Codex, but can also be used manually via its Python API or command-line interface.

Does CryptoGuard require API keys for its security checks?

No, CryptoGuard leverages free public tiers of all integrated security oracles, so no API keys are required for its operation.

Can CryptoGuard guarantee 100% scam prevention?

While CryptoGuard significantly reduces risk by combining multiple security sources, no tool can guarantee 100% detection against novel scam techniques. It serves as a powerful protective layer.

CryptoGuard is an AI agent safety layer that automatically analyzes smart contracts before any crypto transaction, blocking common scams like honeypots, rug pulls, and blacklists.

How does CryptoGuard detect crypto scams?

It integrates multiple independent security oracles (GoPlus, Honeypot.is, TokenSniffer, De.Fi, QuickIntel) and a bytecode scanner to cross-reference potential risks and provide a comprehensive safety assessment.

CryptoGuard: AI Agent Safety for Crypto & DeFi Security

CryptoGuard acts as a crucial pre-transaction safety layer for AI coding agents and manual crypto users alike, diligently analyzing smart contracts before any transaction is executed. It proactively identifies a spectrum of crypto scams, including honeypots, blacklist functions, rug pulls, and malicious tokens. By aggregating insights from over five independent security oracles and performing bytecode analysis, CryptoGuard assigns a risk score to potential interactions, ensuring that high-risk transactions are blocked and your digital assets remain secure.

Características Principales

01Pre-transaction safety hook for AI agents (e.g., Claude Code, OpenAI Codex)

02Integrates multiple independent security oracles (GoPlus, Honeypot.is, TokenSniffer, De.Fi, QuickIntel)

03Provides a Python API and command-line interface for manual checks and programmatic integration

04Configurable risk thresholds (CRITICAL, HIGH, MEDIUM) for automated transaction blocking

05Detects various risks: honeypots, blacklist functions, rug pulls, balance manipulation, extreme taxes, and unverified contracts

061 GitHub stars

Casos de Uso

01Automatically protect AI agents from executing malicious crypto transactions on various chains.

02Manually check the safety and potential risks of any token or smart contract address before interaction.

03Integrate robust security analysis into custom DeFi applications or blockchain-related scripts via Python or MCP server.