Elasticsearch MCP - Full-Text Search & Observability for AI Agents¶
What It Is¶
Elastic MCP exposes Elasticsearch search and observability capabilities over the Model Context Protocol. AI agents can query any index with Elasticsearch Query DSL, run kNN semantic search, and monitor APM traces, logs, and infrastructure metrics.
Tools Available¶
| Tool | Description |
|---|---|
| Full-text search | Query any index with Elasticsearch Query DSL |
| Vector search | kNN semantic search across embeddings |
| Aggregations | Buckets, metrics, pipeline aggregations |
| Observability | APM traces, logs, and infrastructure metrics |
Quick Start¶
npx -y @elastic/mcp-server-elasticsearch
Business Use Cases¶
- Log investigation: "Find all ERROR-level logs from the payments service in the last hour"
- Semantic search: "Find documents related to our refund policy"
- Infrastructure health: "Show me APM traces with latency over 500ms"
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.
Limitations¶
- Elasticsearch 8.x+ required. Vector search needs dense_vector field mappings.