{"id":"gokeep-projects-elasticsearch-mcp-server","name":"elasticsearch-mcp-server","homepage":null,"repo_url":"https://github.com/gokeep-projects/elasticsearch-mcp-server","category":"search","subcategories":[],"tags":["mcp","elasticsearch","quarkus","ai-tools","sse","search","java"],"what_it_does":"Provides a Model Context Protocol (MCP) server that exposes Elasticsearch capabilities as MCP tools, implemented using Quarkus. It starts an SSE endpoint and a streamable HTTP endpoint and forwards requests to a configured Elasticsearch cluster (supports optional basic username/password).","use_cases":["Letting AI agents query an Elasticsearch cluster via standardized MCP tools","Operational introspection (cluster health, index/alias listing, mappings retrieval)","Search workflows using Elasticsearch SQL or DSL through an agent-friendly interface","Retrieval of individual documents by ID"],"not_for":["Public internet deployment without additional access controls","Use cases requiring a fully specified REST/OpenAPI contract for MCP endpoints","Environments needing explicit, documented rate limiting and pagination semantics for MCP tools"],"best_when":"You want a lightweight, locally deployable MCP bridge from agent workflows to Elasticsearch operations.","avoid_when":"You require strict governance features (auditing, fine-grained auth, documented quotas) or you plan to expose the service publicly without network-level protections.","alternatives":["Use Elasticsearch’s native APIs directly from the agent (custom tool wrappers)","Other MCP servers for Elasticsearch or OpenSearch (if available in your ecosystem)","A generic proxy/gateway that translates MCP tool calls to Elasticsearch REST calls"],"af_score":45.8,"security_score":39.5,"reliability_score":23.8,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-04-04T20:01:48.468233+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":"http://{ip}:19000/mcp (streamable) and http://{ip}:19000/mcp/sse (SSE)","has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["Elasticsearch basic authentication via username/password (configurable for connecting to Elasticsearch)"],"oauth":false,"scopes":false,"notes":"README describes connecting to Elasticsearch with optional username/password, but does not document authentication/authorization for the MCP server endpoints themselves."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"Appears to be an open-source project (MIT) intended for self-hosting; no hosted pricing information provided."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":45.8,"security_score":39.5,"reliability_score":23.8,"mcp_server_quality":70.0,"documentation_accuracy":55.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":75.0,"rate_limit_clarity":10.0,"tls_enforcement":40.0,"auth_strength":35.0,"scope_granularity":10.0,"dependency_hygiene":45.0,"secret_handling":70.0,"security_notes":"TLS requirement for MCP endpoints is not documented; authentication for MCP endpoints is not described (only Elasticsearch credentials for upstream access). Secrets appear intended to be provided via CLI args/env vars, but guidance about logging/redaction is not documented. Since the service can connect to Elasticsearch with credentials, ensure proper network controls and avoid exposing the MCP server publicly without additional safeguards.","uptime_documented":0.0,"version_stability":35.0,"breaking_changes_history":40.0,"error_recovery":20.0,"idempotency_support":"false","idempotency_notes":"No explicit statement about idempotency for tools; many operations are read-only in the listed toolset, but tool list and semantics for write operations are not documented.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Default MCP server listens on 0.0.0.0:19000 and auto-enables SSE and streamable endpoints; ensure network exposure is controlled.","Default Elasticsearch connection is http://localhost:9200 with no password; misconfiguration may lead to accidental connections to unintended clusters.","Only communication endpoints are documented; tool schemas, argument validation behavior, and error payload formats are not shown in the provided README."]}}