Elastic documents the recommended Agent Builder Model Context Protocol endpoint at `{KIBANA_URL}/api/agent_builder/mcp` (or `{KIBANA_URL}/s/{SPACE_NAME}/api/agent_builder/mcp` for custom Kibana spaces) per elastic.co/docs/explore-analyze/ai-features/agent-builder/mcp-server. The MCP server exposes built-in and custom Agent Builder tools to Claude Desktop, Cursor, VS Code, and other MCP clients via `npx mcp-remote` with an `Authorization: ApiKey` header. API keys must include the Kibana application privilege `feature_agentBuilder.read` or clients receive HTTP 403. Elastic notes the legacy `elastic/mcp-server-elasticsearch` project is deprecated in favor of this endpoint on Elastic 9.2+ and Elasticsearch Serverless; docs recommend least-privilege index scopes and API key expiration.
Use cases
- Let Cursor query observability or security indices through governed Agent Builder tools
- Expose only `logs-*` and `metrics-*` indices via scoped API keys in production
- Connect MCP clients without maintaining the deprecated mcp-server-elasticsearch container
- Use custom Kibana space URLs when tools are space-scoped
- Pair MCP access with Elastic AI features documented for 9.2+ deployments
Key features
- Cursor
- Claude Desktop
- VS Code
- Codex
Frequently Asked Questions
- What happened to elastic/mcp-server-elasticsearch?
- Elastic docs deprecate that GitHub project in favor of the Agent Builder MCP endpoint on 9.2+ and Serverless.
- Why am I getting HTTP 403?
- Docs state the API key must include Kibana application privilege feature_agentBuilder.read.
- Can I restrict which indices agents see?
- Yes—Elastic best-practice examples limit API key index privileges to required patterns such as logs-* and metrics-*.
Related
Related
3 Indexed items
Algolia Productivity MCP Server
Algolia documents an official managed Model Context Protocol server at algolia.com/doc/guides/model-context-protocol/productivity-mcp. Connect MCP clients to the remote HTTP endpoint `https://mcp.algolia.com/mcp` with OAuth (enable under Generate AI in the Algolia dashboard; sign in when prompted so the MCP inherits your account permissions). Productivity MCP is user-scoped and read-only per docs—tools cover search (`algolia_search_list_indices`, `algolia_search_index`, `algolia_search_for_facet_values`), Recommend (`algolia_recommendations`), and analytics helpers such as top searches, no-click rates, filter usage, and user counts. Algolia docs distinguish this from Algolia Public MCP for application-scoped, curated index exposure to external agents. Supported clients include ChatGPT, Claude, Claude Code, Cursor, Gemini CLI, VS Code, and OpenAI Playground.
Jina AI MCP Server
Jina AI documents an official remote Model Context Protocol server in the jina-ai/MCP repository at https://mcp.jina.ai/v1 using Streamable HTTP transport (MCP spec 2025-03-26). Tools expose Jina Reader, Embeddings, and Reranker APIs: primer for contextual status; read_url and parallel_read_url for URL-to-markdown extraction; capture_screenshot_url and guess_datetime_url for page screenshots and publish-date hints; search_web, search_arxiv, search_ssrn, and search_images for web and specialized search; expand_query for query rewriting; sort_by_relevance for reranking; classify_text and deduplicate_strings for embedding-powered text tasks per the README. Clients with native remote MCP support connect directly with Authorization Bearer JINA_API_KEY; others use npx mcp-remote https://mcp.jina.ai/v1. Optional URL filters include_tags/exclude_tags to trim tool lists. Pairs with the jina-ai tool entry for agent-driven reading and search workflows.
Meilisearch MCP Server
Meilisearch maintains an official Model Context Protocol server in meilisearch/meilisearch-mcp, documented at meilisearch.com/blog/introducing-mcp-server. The Python stdio server connects MCP clients to any running Meilisearch instance via `MEILI_HTTP_ADDR` and optional `MEILI_MASTER_KEY`, with `update-connection-settings` to switch hosts mid-session. Tools cover index management, document ingestion, search (filters, sorting, facets, semantic/hybrid), settings, API keys, tasks, and health checks per the README. Install paths include `uvx meilisearch-mcp`, pip, source, and Docker (`getmeili/meilisearch-mcp`). Meilisearch notes the server is development-oriented and that native Meilisearch MCP transport support is coming.