onchain-intel - AI-powered on-chain intelligence for Ethereum addresses. Analyze contracts, balances, transactions, and token activity from the terminal in seconds.

onchain-intel is an open-source CLI tool for turning an Ethereum address into structured on-chain intelligence. It combines Etherscan data with an LLM-driven analysis to help users quickly understand whether an address is a verified contract, a proxy, a wallet, or an active participant in the ecosystem.

It is designed for security reviews, wallet research, protocol due diligence, and demo-ready storytelling. The output is structured and easy to share in pitch decks, internal reports, or developer workflows.

What it does

Given an Ethereum address, onchain-intel can:

• inspect contract metadata such as verification status, compiler details, proxy status, and implementation address
• report native ETH balance
• summarize recent transactions and account activity
• surface ERC-20 transfer activity and token-related behavior
• generate a structured intelligence report with contract type, risk flags, activity summary, and notable findings

Core experience

The tool supports two analysis modes:

• Agent mode: a single-agent workflow that uses the available on-chain tools to answer a prompt and return a structured report
• Graph mode: a multi-step workflow with separate researcher, analyst, and reporter stages for richer analysis

It also supports conversational continuity through thread-based memory, so follow-up questions can stay grounded in the same context.

Quick start

1. Install dependencies:

   uv sync

2. Set the required environment variables:

   export ETHERSCAN_API_KEY=your_etherscan_key
   export GROQ_API_KEY=your_groq_key

3. Run the analyzer against any Ethereum address:

   uv run python -m onchain_intel 0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48

4. Use graph mode for a more detailed workflow:

   uv run python -m onchain_intel 0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48 --mode graph --verbose

Example output

The tool returns a structured JSON report with fields such as:

• address
• summary
• contract_type
• is_verified
• is_proxy
• risk_flags
• recent_activity_summary
• notable_findings

Why it is useful

• fast way to turn a raw address into an interpretable on-chain profile
• useful for analysts, builders, and teams evaluating contracts or wallet behavior
• easy to demonstrate in a live session because it produces clear, structured outputs
• open-source and suitable for public showcases, product demos, and marketing narratives

Project structure

• src/onchain_intel/agent.py — single-agent workflow
• src/onchain_intel/graph.py — multi-step research/analysis/reporting workflow
• src/onchain_intel/etherscan.py — Etherscan data access layer
• src/onchain_intel/tool.py — tool wrappers exposed to the agent
• src/onchain_intel/schema.py — structured report schema

License

This project is open source and intended for community use and demonstration.