List loans. Monitor expiry — silent auto-liquidation.
AI agents call get_loans to retrieve information from Basis MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The primary function is read-only: listing and retrieving loan data. However, the contextual information about silent auto-liquidation and expiry monitoring elevates severity from 'low' to 'high' because misuse (e.g., an agent ignoring liquidation warnings) could lead to financial loss.
From the tool's definition 'List loans' — retrieves loan data without modifying or deleting it. However, the description mentions 'Monitor expiry — silent auto-liquidation', indicating the tool surfaces critical state information about loans at risk of liquidation.
Attacks that exploit this kind of access
List loans. Monitor expiry — silent auto-liquidation. It is categorised as a Read tool in the Basis MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Basis MCP Server MCP server in PolicyLayer and add a rule for get_loans: 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 Basis MCP Server. Nothing to install.
get_loans 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_loans 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_loans. 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_loans is provided by the Basis MCP Server MCP server (launch-on-basis/mcp-ts). 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 →