Indexes and documents of an Azure AI Search service: full-text, filtered and vector search, documents by key. Document writes and deletes wait for approval.
The screenshots show GateRook 2.0.0: the Azure AI Search form, the endpoint selector, and this connector’s actual tool preview. The form and preview have not been saved. Select a screenshot to open it at full size.
1. Create a connection and enter the settings
In GateRook, open Connections, select Add connection, search for Azure AI Search, choose its card and select Continue to configure.
Copy the endpoint of the Azure AI Search service. Decide whether to authenticate with an API key or Microsoft Entra ID.
A query key can search and read documents but cannot list or describe indexes. With Entra ID, enable the service’s role-based access and grant the identity the required data-plane roles.
Enter a Name, for example azure-ai-search-demo. Your AI app uses this name to choose the connection.
Enter Service endpoint and restrict Indexes to the intended index names or patterns.
Choose API key under Sign in with and enter a query key for a search-only workflow. Use an admin key only if you deliberately need its broader capabilities.
For Microsoft Entra ID, select the displayed credential source. Supply Tenant (directory) ID and the service-principal fields when required, or use the Azure CLI/browser sign-in flow.
Choose Development, Staging or Production to match the service. Review the connector’s limit and access settings before continuing.
2. Test the connection and add it to an endpoint
Select Continue to test. GateRook checks the settings from your computer. Review the result and resolve any connection, certificate or permission error before continuing.
A query key can be valid even when index-listing tools are denied. Check a named index with a search before broadening permissions. With Entra ID, verify the service permits role-based access and that role assignments have taken effect.
After the test passes, select Continue to add to an endpoint. Choose the endpoint your AI app uses, review its tool modes, then select Save connection. GateRook offers default as the starter endpoint.
3. Use the connection from your AI IDE
After saving Azure AI Search on the endpoint and passing its connection test, ask your AI app to use azure-ai-search-demo, or the name you chose. If its tools are missing, confirm the endpoint assignment and refresh the MCP connection or start a new chat.
The preview below shows this connector’s default tools before saving. The table explains what each tool can do; connection settings, credentials and endpoint policies can restrict access further.
Start with a small read request. Replace example resource names with existing resources your connection may access. Open Activity after the call to inspect its connection, tool, arguments, policy decision and result metadata.
Ask your AI, for example
- “Search the products index on azure-ai-search-demo for wireless headphones and return five matches.”
- “Using azure-ai-search-demo, fetch product-123 from products.”
- “Search products on azure-ai-search-demo for in-stock accessories without changing the index.”
| Tool | What it does | Starts as |
|---|---|---|
search_list_indexes | List the permitted Azure AI Search indexes when the credentials allow index discovery. | Read only |
search_describe_index | Show readable field definitions, search/filter/sort/facet flags, vector dimensions, semantic search configuration and available index statistics. Definition access depends on the credentials. | Read only |
search_query | Search text or a configured vector field with OData filters, facets, sorting and pagination. Vector search requires a suitable vector field. | Read only |
search_get_documents | Documents by id, up to 100. | Read only |
search_index_documents | Adds or replaces up to 100 documents of one index, or merges their fields into existing ones. You approve the exact documents. Never retried. | Approval |
search_delete_documents | Deletes documents by id, up to 1,000 of one index. You approve the exact ids; nothing else is deleted. Never retried. | Approval |
You can change each tool’s mode on the endpoint. Every call your AI makes appears in Activity.