OpenSearch documents an open-source Model Context Protocol server at docs.opensearch.org/latest/ai-agent-integrations/mcp-server for AI assistants to interact with OpenSearch clusters via MCP tools instead of raw REST. The opensearch-project/opensearch-mcp-server-py package supports stdio (Claude Desktop, Cursor, Kiro) and streaming transports (SSE/Streamable HTTP) with tools for listing indexes, retrieving mappings, running search queries, checking cluster health, and counting documents per docs. Configure single-cluster mode via environment variables or multi-cluster YAML; authentication supports basic auth, IAM, header auth, and mTLS for self-managed OpenSearch, Amazon OpenSearch Service, and Serverless. OpenSearch 3.0+ also ships an experimental in-cluster MCP endpoint at `/_plugins/_ml/mcp` (Streamable HTTP) per ML Commons docs—distinct from the standalone py server for external clients.
Use cases
- Let agents search OpenSearch indexes and inspect mappings from Cursor or Claude Desktop
- Prototype agent analytics over Amazon OpenSearch Service without hand-written DSL
- Compare standalone opensearch-mcp-server-py against built-in OpenSearch 3.0 /_plugins/_ml/mcp
- Pair with voyage-ai or vespa entries when evaluating hybrid RAG retrieval stacks
- Use multi-cluster YAML to scope agent access across staging and production indices
Key features
- Claude Desktop
- Cursor
- Kiro
- LangChain
Frequently Asked Questions
- Is this the same as OpenSearch 3.0 built-in MCP?
- OpenSearch docs distinguish the standalone opensearch-mcp-server-py package from the experimental in-cluster /_plugins/_ml/mcp endpoint in OpenSearch 3.0+.
- Does it work with Amazon OpenSearch Serverless?
- OpenSearch MCP docs list compatibility with self-managed clusters, Amazon OpenSearch Service, and Serverless collections.
- Which transports are supported?
- Docs describe stdio for desktop clients and streamable HTTP/SSE for remote or agent-framework deployments.
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.
Elastic Agent Builder MCP Server
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.
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.