{"id":"cdatasoftware-cdata-jdbc-mcp-server","name":"cdata-jdbc-mcp-server","homepage":"https://www.cdata.com/solutions/mcp","repo_url":"https://github.com/CDataSoftware/cdata-jdbc-mcp-server","category":"ai-ml","subcategories":[],"tags":["mcp","stdio","jdbc","data-integration","ai-agent","desktop-integration","read-only"],"what_it_does":"Provides a generic MCP (Model Context Protocol) server that uses CData JDBC drivers to expose external data sources as SQL-like tables/columns and lets MCP clients run queries (and, despite the repo name/readme emphasis, it claims CRUD tools are available via MCP tools). Communication with the MCP client is via stdio on the same machine.","use_cases":["Ask natural-language questions about live data from hundreds of SaaS/DB/API sources without writing SQL","Retrieve table/column metadata for a connected data source via MCP tools","Run read-oriented SQL SELECT queries against CData JDBC driver-backed sources","Prototype agent-driven reporting/analytics workflows for business data (e.g., CRM opportunities, tickets, calendar events)"],"not_for":["Internet-exposed or multi-host deployments that require networked isolation (stdio-only implies local-only use)","Use cases requiring robust API server features like pagination contracts, durable job queues, or centralized rate-limit handling","Security-sensitive environments that require strict guarantees around query safety/row-level controls (not described)"],"best_when":"When you run the MCP server locally alongside an MCP-capable desktop client (e.g., Claude Desktop) and you want agent-accessible, near-real-time data retrieval across many JDBC-supported sources.","avoid_when":"Avoid when you need a hosted/public API endpoint, cross-machine access, strong auth/rbac for the MCP server itself, or documented query-safety controls (since the tool exposes SQL execution).","alternatives":["CData’s other MCP servers (beta) if they provide simplified setup or broader capabilities","Direct use of CData JDBC drivers from your own application (no MCP layer)","Other MCP server implementations tailored to specific data providers","Building a custom MCP server that wraps a curated data access layer rather than exposing raw SQL"],"af_score":45.5,"security_score":27.0,"reliability_score":27.5,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-04-04T19:41:25.592737+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 <driver>.jar --license)","Data-source authentication via the JDBC driver (e.g., OAuth in browser when required) as part of the JDBC connection string/JdbcUrl used in the .prp file"],"oauth":false,"scopes":false,"notes":"The MCP server itself appears to rely on the JDBC driver’s authentication configured in the .prp/JdbcUrl (including possible OAuth flows). The README does not describe additional auth between MCP client and server."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":["CData JDBC driver licensing is required per data source (implied by license step in README)","CData MCP Server (beta) is mentioned for full read/write capabilities (pricing not provided here)"],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"No explicit MCP server pricing is provided in the README; CData JDBC driver licensing is required to connect to sources."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":45.5,"security_score":27.0,"reliability_score":27.5,"mcp_server_quality":78.0,"documentation_accuracy":70.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":45.0,"rate_limit_clarity":0.0,"tls_enforcement":10.0,"auth_strength":30.0,"scope_granularity":10.0,"dependency_hygiene":50.0,"secret_handling":40.0,"security_notes":"Security controls for the MCP layer are not described (no auth/authorization between MCP client and server). The server runs locally via stdio, which reduces network exposure but not local privilege risk. Authentication appears to be handled by the underlying CData JDBC driver (including OAuth if applicable). The presence of a SQL execution tool increases risk of overly broad data access or expensive queries; the README does not document query restrictions, row-level security, audit logging, or secret-handling practices.","uptime_documented":0.0,"version_stability":40.0,"breaking_changes_history":40.0,"error_recovery":30.0,"idempotency_support":"false","idempotency_notes":null,"pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Server uses stdio, so it can only be used by clients running on the same machine.","The README claims read-only repo name but also states the AI client can use built-in tools to read/write/update/delete; tool exposure may not match “read-only” expectations—agents should be conservative.","SQL execution tool ({servername}_run_query) can be powerful; the README does not document safeguards, query parameterization, or restrictions against expensive/unsafe queries.","Metadata/tools output is returned as CSV; agents may need to parse CSV and handle quoting/encoding."]}}