AI agents call searchEndpoints to retrieve information from Swagger without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Given the context that this MCP server wraps Swagger/OpenAPI specifications to enable AI assistants to 'query API endpoints, schemas, and generate code examples,' searchEndpoints appears to be a read-only discovery tool. While the description is absent (lowering confidence slightly), the function name and the pattern of sibling tools all point to information retrieval.
From the tool's definition Tool name is 'searchEndpoints' which indicates querying/searching API endpoints. The description is empty, but the sibling tools (getApiDocumentation, getEndpointCategories, getExample, getSchema, test) are all Read operations, suggesting this server is…
Documented attack patterns abuse exactly the kind of access searchEndpoints gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Swagger, and nothing reaches the server without passing your rules. This is the rule we recommend for searchEndpoints:
{
"version": "1",
"default": "deny",
"tools": {
"searchEndpoints": {}
}
}searchEndpoints is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
searchEndpoints. It is categorised as a Read tool in the Swagger MCP Server, which means it retrieves data without modifying state.
Register the Swagger MCP server in PolicyLayer and add a rule for searchEndpoints: 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 Swagger. Nothing to install.
searchEndpoints 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 searchEndpoints 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 searchEndpoints. 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.
searchEndpoints is provided by the Swagger MCP server (salacoste/openapi-mcp-swagger). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Swagger, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
6 Swagger tools catalogued and risk-classified — across an index of 43,000+ MCP servers.