{"id":"cdatasoftware-confluence-mcp-server-by-cdata","name":"confluence-mcp-server-by-cdata","homepage":null,"repo_url":"https://github.com/CDataSoftware/confluence-mcp-server-by-cdata","category":"communication","subcategories":[],"tags":["mcp","confluence","cdata","jdbc","stdio","data-access","read-only","local-connector"],"what_it_does":"Provides a local, read-only (per README) MCP server that exposes Confluence data via CData’s JDBC Driver as MCP tools. It runs as a stdio process (suitable for clients like Claude Desktop on the same machine) and supports tool-based table discovery plus SQL SELECT queries through a JSON-RPC tool interface.","use_cases":["Ask natural-language questions about Confluence content/metadata via an MCP-capable desktop client","Use MCP tools to list available tables/columns and run read-only SQL SELECT queries against Confluence (through the CData JDBC model layer)","Build local AI-assisted analytics/workflows over Confluence data without writing SQL (where the client can translate questions)"],"not_for":["High-throughput or remote/network MCP usage (stdio implies local same-host use)","Public/hosted multi-tenant deployments without additional hardening","Use cases requiring guaranteed data integrity for writes/updates/deletes (README positions this project as read-only)","Environments needing an explicit REST/HTTP service contract or managed API gateway"],"best_when":"You have a local MCP client (e.g., Claude Desktop) and want an on-machine connector to query live Confluence data using CData’s JDBC mapping.","avoid_when":"You need a deployable remote API endpoint, strong documented operational guarantees (SLA, rate limits), or explicit write capabilities beyond what the connector documents.","alternatives":["CData’s managed “Connect AI” MCP platform (remote) mentioned in README","CData’s full MCP Server for Confluence (for CRUD) referenced in README","Direct use of CData JDBC driver from your own application (no MCP)"],"af_score":44.5,"security_score":37.8,"reliability_score":20.0,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-04-04T19:59:29.222623+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 ... --license)","JDBC connection authentication via connection-string/JdbcUrl (may include OAuth flows per README note)"],"oauth":true,"scopes":false,"notes":"Authentication is handled by the underlying CData JDBC driver using the provided JDBC URL/connection string; README notes OAuth may require browser-based login. The MCP server itself is started locally with a .prp file pointing to the driver and JDBC URL."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":["CData JDBC Driver licensing required (trial/license key mentioned)"],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"README indicates a JDBC driver license/trial mechanism, but does not provide pricing tiers or credit-card requirements for the MCP server itself."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":44.5,"security_score":37.8,"reliability_score":20.0,"mcp_server_quality":55.0,"documentation_accuracy":65.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":65.0,"rate_limit_clarity":0.0,"tls_enforcement":20.0,"auth_strength":55.0,"scope_granularity":20.0,"dependency_hygiene":40.0,"secret_handling":50.0,"security_notes":"Security controls are largely delegated to the underlying CData JDBC driver and the Confluence/OAuth mechanisms configured in JdbcUrl. The MCP server runs locally via stdio, which reduces exposure over the network but increases risk if the host is compromised. README does not describe TLS enforcement explicitly (likely depends on driver/JdbcUrl), does not mention fine-grained scopes/authorization boundaries within MCP, and provides no guidance on how secrets in .prp/JdbcUrl are stored/handled (though typical practice is local file-based config). Dependency/version hygiene is not verifiable from provided content.","uptime_documented":0.0,"version_stability":30.0,"breaking_changes_history":20.0,"error_recovery":30.0,"idempotency_support":"false","idempotency_notes":"Read-only tools are described (get_tables/get_columns/run_query with SELECT). Idempotency for any write behavior is not documented for this project; README positions it as read-only, so idempotency is likely implicit for SELECT-only usage but not explicitly guaranteed.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["MCP server uses stdio, so it can only be used with clients running on the same machine","Tool outputs are returned in CSV format for get_columns/get_tables; agents may need to parse CSV robustly","No explicit guidance on rate limits, query size limits, or error codes is provided in the README","The .prp configuration must include correct DriverPath/DriverClass/JdbcUrl and optional Tables; misconfiguration can prevent tools from appearing in the client"]}}