AI agents call eth_get_balance to retrieve information from PortalMCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries blockchain state to retrieve balance information. It performs no writes, deletions, executions, or financial transactions. Even though it operates on a financial system (Ethereum), it is purely informational—reading balances without moving funds. The blast radius of misuse is minimal; an attacker could enumerate balances but cannot steal funds or modify state.
From the tool's definition Tool name 'eth_get_balance' and description 'Get ETH balance for an Ethereum address or the default signer wallet' indicate a retrieval operation with no side effects.
Attacks that exploit this kind of access
Get ETH balance for an Ethereum address or the default signer wallet. It is categorised as a Read tool in the PortalMCP MCP Server, which means it retrieves data without modifying state.
Register the Portal MCP server in PolicyLayer and add a rule for eth_get_balance: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches PortalMCP. Nothing to install.
eth_get_balance is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the eth_get_balance rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for eth_get_balance. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
eth_get_balance is provided by the Portal MCP server (portalfnd/portalmcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →