EVMSAFE
MCP server that provides LLMs with tools for interacting with EVM networks
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
A comprehensive Model Context Protocol (MCP) server that provides blockchain services across 60+ EVM-compatible networks. This server enables AI agents to interact with Ethereum, Optimism, Arbitrum, Base, Polygon, and many other EVM chains with a unified interface through 22 tools and 10 AI-guided prompts.
📋 Contents
- Overview
- Features
- Supported Networks
- Prerequisites
- Installation
- Configuration
- Environment Variables
- Server Configuration
- Usage
- API Reference
- Tools
- Prompts
- Resources
- Security Considerations
- Project Structure
- Development
- License
🔭 Overview
The MCP EVM Server leverages the Model Context Protocol to provide blockchain services to AI agents. It supports a wide range of services including:
- Reading blockchain state (balances, transactions, blocks, etc.)
- Interacting with smart contracts with automatic ABI fetching from block explorers
- Transferring tokens (native, ERC20, ERC721, ERC1155)
- Querying token metadata and balances
- Chain-specific services across 60+ EVM networks (34 mainnets + 26 testnets)
- ENS name resolution for all address parameters (use human-readable names like 'vitalik.eth' instead of addresses)
- AI-friendly prompts that guide agents through complex workflows
All services are exposed through a consistent interface of MCP tools, resources, and prompts, maki
49b4203eb85aOBSERVED · 2026-10-02Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add evm-mcp-server --env ETHERSCAN_API_KEY=${ETHERSCAN_API_KEY} --env EVM_PRIVATE_KEY=${EVM_PRIVATE_KEY} -- npx -y @mcpdotdirect/[email protected]{
"mcpServers": {
"evm-mcp-server": {
"command": "npx",
"args": [
"-y",
"@mcpdotdirect/[email protected]"
],
"env": {
"ETHERSCAN_API_KEY": "${ETHERSCAN_API_KEY}",
"EVM_PRIVATE_KEY": "${EVM_PRIVATE_KEY}"
}
}
}
}Exposed tools (25)
22 read · 3 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
approve_token_spending | read | |
get_allowance | read | |
get_balance | read | |
get_block | read | |
get_chain_info | read | |
get_contract_abi | read | |
get_erc1155_balance | read | |
get_gas_price | read | |
get_latest_block | read | |
get_nft_info | read | |
get_supported_networks | read | |
get_token_balance | read | |
get_transaction | read | |
get_transaction_receipt | read | |
get_wallet_address | read | |
lookup_ens_address | read | |
multicall | read | |
read_contract | read | |
resolve_ens_name | read | |
sign_message | read | |
sign_typed_data | read | |
transfer_erc20 | write | |
transfer_native | write | |
wait_for_transaction | read | |
write_contract | write |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (1 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (1)
@modelcontextprotocol/sdk, express, viem, zod, @types/bun, @types/express, @types/node, conventional-changelog-cli
Gates applied: no_behavioural_pass.
49b4203eb85afull audit observations/trust-audit/mcp-server/mcpdotdirect__evm-1.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-02 | 49b4203eb85a | SAFE | B | 89 | first audit |
Questions
What is the EVM MCP server?
MCP server that provides LLMs with tools for interacting with EVM networks
What tools does EVM expose?
25 in total: 22 read-only, 3 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is EVM safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does EVM need?
It reads ETHERSCAN_API_KEY and EVM_PRIVATE_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does EVM run?
It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on npm as @mcpdotdirect/evm-mcp-server at 2.0.4.
How current is this page?
The grade is for one exact copy of the source (49b4203eb85a), read on 2026-10-02. The repository is watched and re-audited when it changes.