{"id":"cdatasoftware-apache-hadoop-hive-mcp-server-by-cdata","name":"apache-hadoop-hive-mcp-server-by-cdata","homepage":null,"repo_url":"https://github.com/CDataSoftware/apache-hadoop-hive-mcp-server-by-cdata","category":"ai-ml","subcategories":[],"tags":["mcp","stdio","data-connector","hadoop","hive","jdbc","llm-tools","read-only","java","local-integration"],"what_it_does":"Provides a local, read-only Model Context Protocol (MCP) server that exposes Apache Hadoop Hive data via CData JDBC Driver as MCP tools (get tables, get columns, run SELECT queries), so local LLM clients (e.g., Claude Desktop) can query live Hive data without writing SQL.","use_cases":["Ask natural-language questions over live Hive tables/columns","Discover available Hive tables and schemas (columns) via MCP tools","Run read-only analytical queries (SELECT) against Hive through an LLM tool workflow"],"not_for":["Direct public/remote access to Hive over a network (server is stdio/local)","High-assurance transactional workloads or write-heavy CRUD scenarios (README positions this project as read-only)","Use-cases requiring strict pagination, data export governance, or fine-grained query result limits (not described)"],"best_when":"You have Hive connectivity via the CData JDBC Driver, and you want a local MCP bridge for an on-device desktop LLM client to run read-only queries.","avoid_when":"You need robust production-grade API semantics (auth scopes, rate-limit headers, pagination contracts, structured error codes) or remote multi-tenant access.","alternatives":["CData managed MCP server offering (CData Connect AI / CData MCP Server for Apache Hadoop Hive)","Use Hive directly with an application/service layer and an internal API","Build/host an MCP server that fronts a dedicated read-only query service with stronger governance"],"af_score":42.2,"security_score":34.8,"reliability_score":12.5,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-04-04T19:49:14.688133+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":["CData JDBC Driver licensing (java -jar cdata.jdbc... --license)","Underlying CData JDBC connection credentials (may include OAuth depending on the JDBC driver configuration)"],"oauth":true,"scopes":false,"notes":"No dedicated MCP auth layer is described; authentication is delegated to the CData JDBC Driver connection string/JDBC driver configuration."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"README describes licensing of the CData JDBC driver; this repo itself is MIT-licensed but pricing for connectivity depends on CData driver/managed offerings."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":42.2,"security_score":34.8,"reliability_score":12.5,"mcp_server_quality":62.0,"documentation_accuracy":70.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":55.0,"rate_limit_clarity":0.0,"tls_enforcement":40.0,"auth_strength":40.0,"scope_granularity":20.0,"dependency_hygiene":45.0,"secret_handling":30.0,"security_notes":"Security controls are largely not described at the MCP layer. Authentication is delegated to CData JDBC Driver configuration (potentially OAuth), but there is no mention of OAuth scope granularity, MCP-level access control, audit logging, query constraints, or TLS/transport behavior for Hive connectivity. The stdio/local design reduces some network exposure but credentials may still be stored in the .prp/JDBC connection string on the host.","uptime_documented":0.0,"version_stability":20.0,"breaking_changes_history":20.0,"error_recovery":10.0,"idempotency_support":"false","idempotency_notes":null,"pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Server uses stdio; it must run on the same machine as the MCP client.","README lists run_query as SELECT; agents should avoid attempting non-SELECT/CRUD operations since the project is positioned as read-only.","Large result sets may be returned as CSV without documented limits/pagination, which can overwhelm client context."]}}