ChatOps Using OpenSearch MCP Server
summary
Demo of opensearch-mcp-server-py connects OpenSearch clusters to AI CLIs like Kiro CLI and runs concrete ops checks: cluster health, cat allocations, index mappings, query insights for expensive queries, and slow log toggles. It generates actionable fixes (disable full‑text fields, enable global ordinals, create new index + reindex) and a memory-backed agent store, but sometimes misinterprets host context (e.g., Linux RAM) so recommendations require validation and use of read‑only MCP credentials to avoid destructive changes.
AI-written summary · about
chapters (10)
- 0:00OpenSearch MCP Server intro
- 3:34Demo with Kiru CLI and MCP
- 5:00Operational cluster reviews
- 7:47Best practices analysis
- 13:28Index mapping optimizations
- 17:24Debugging slow queries
- 21:01Recommendations validation
- 23:07Preventing data changes
- 24:53MCP server best practices
- 27:44LLM accuracy with context
related talks
37mAgentic AI Meets Search: Real-World Use Cases With...
33mAccelerating OpenSearch Service Operations With an Operation Agent
36mBuilding Next-gen Search With OpenSearch
16mOne AI Chat to Rule Them All: Connecting Enterprise Tools with MCP
28mTalk to your data without coding agentic AI and Claude with the MCP protocol
21mNoOps for OpenSearch: The Future of Self-Healing Search Infras...
27mAlert Fatigue to Action: Observability as the Backbone of Practical AIOps with OpenSearch
36mPutting AI Agents to Work: A New Era of Open Source Connectivity with MCP