AI agents call get_token_holdings to retrieve information from Onchain without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries existing blockchain data without creating, modifying, deleting, or executing any state-changing operations. It is purely informational, consistent with the Read category. The blast radius is minimal—an AI agent can only retrieve information about token holdings, not affect them. Low severity is appropriate for read-only blockchain data queries.
From the tool's definition Tool description explicitly states 'List the non-zero SPL token balances held by a Solana wallet' — a read-only query operation with no side effects.
Attacks that exploit this kind of access
List the non-zero SPL token balances held by a Solana wallet. It is categorised as a Read tool in the Onchain MCP Server, which means it retrieves data without modifying state.
Register the Onchain MCP server in PolicyLayer and add a rule for get_token_holdings: 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 Onchain. Nothing to install.
get_token_holdings 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 get_token_holdings 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 get_token_holdings. 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.
get_token_holdings is provided by the Onchain MCP server (sylvainlondon136/onchain-mcp). 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 →