{"id":"cdatasoftware-google-search-mcp-server-by-cdata","name":"google-search-mcp-server-by-cdata","homepage":null,"repo_url":"https://github.com/CDataSoftware/google-search-mcp-server-by-cdata","category":"ai-ml","subcategories":[],"tags":["mcp","stdio","cdata","jdbc","google-search","read-only","desktop-integration"],"what_it_does":"Provides a local, read-only MCP server that exposes Google Search data as relational tables via CData's JDBC Driver. The MCP server runs over stdio so an MCP-capable desktop client (e.g., Claude Desktop) can query live Google Search data using tool calls like get_tables, get_columns, and run_query (SQL SELECT).","use_cases":["Letting an LLM answer questions about live Google Search data without writing SQL","Building desktop-agent workflows that need read-only access to Google Search-derived datasets","Querying/searching structured results through a table/column abstraction backed by CData JDBC"],"not_for":["Browser-based or remote MCP access (requires the client to run on the same machine due to stdio transport)","Write/update/delete operations (the project is described as read-only)","High-throughput or service-style API usage where rate-limit and pagination controls are required"],"best_when":"You want a local MCP integration on the same host as your client, and you need read-only Google Search data access via a JDBC-to-table model.","avoid_when":"You need a network-accessible MCP endpoint, fine-grained API rate-limit/pagination semantics, or robust server-side error codes tailored to MCP tool calls.","alternatives":["Use CData’s managed MCP platform (CData Connect AI) referenced in the README","Use CData’s non-MCP JDBC driver directly from your own application","If available, use a dedicated Google Search API integration rather than a JDBC-to-SQL abstraction"],"af_score":49.8,"security_score":34.5,"reliability_score":25.0,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-04-04T19:59:07.032682+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":null,"has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["Configured via CData JDBC Driver connection string / OAuth flow (as applicable to the driver)"],"oauth":true,"scopes":false,"notes":"Authentication is performed through the underlying CData JDBC driver configuration (the README notes that OAuth may be required and must be authenticated in a browser). The MCP server itself appears to rely entirely on the JDBC driver credential configuration in the .prp file."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"The README does not describe pricing for this MCP server; it references licensing for the CData JDBC Driver and mentions a managed platform (CData Connect AI)."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":49.8,"security_score":34.5,"reliability_score":25.0,"mcp_server_quality":70.0,"documentation_accuracy":65.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":50.0,"rate_limit_clarity":5.0,"tls_enforcement":20.0,"auth_strength":55.0,"scope_granularity":10.0,"dependency_hygiene":45.0,"secret_handling":40.0,"security_notes":"The MCP server is local via stdio, reducing exposure to network interception, but the README does not document transport hardening, credential storage practices, or secret redaction in logs. Authentication is delegated to the underlying CData JDBC driver configuration (potentially OAuth), but the README does not describe fine-grained scopes or least-privilege controls. No dependency or vulnerability hygiene information is provided in the supplied content.","uptime_documented":0.0,"version_stability":40.0,"breaking_changes_history":30.0,"error_recovery":30.0,"idempotency_support":"false","idempotency_notes":"Read-only operations are expected (tools listed are get_tables/get_columns and run_query with SELECT). However, the README does not specify idempotency semantics or detailed behavior for retries/timeouts.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Server uses stdio transport; the client must run on the same machine as the MCP server.","May require fully exiting/reopening the client (Claude Desktop) for MCP servers to appear.","Tool outputs are returned in CSV (first row headers); agents must parse accordingly.","Requires correct JDBC driver licensing and correct connection string in the .prp file; otherwise tool calls will fail."]}}