manticore
Manticore Search MCP Server
claude mcp add --transport stdio krajcik-manticore-mcp-server /path/to/manticore-mcp-server \ --env DEBUG="false" \ --env MANTICORE_URL="http://localhost:9308" \ --env REQUEST_TIMEOUT="30s" \ --env MAX_RESULTS_PER_QUERY="50"
How to use
The Manticore MCP Server exposes a set of MCP-compatible tools that bridge Manticore Search functionality to clients (such as Claude Code, Cursor, and other AI development tools). It provides capabilities for full-text search, index/table management, document insertion and updates, as well as cluster management and status checks. Clients can discover and invoke tools like search, show_tables, describe_table, insert_document, update_document, delete_document, show_cluster_status, and cluster management operations (create_cluster, join_cluster, alter_cluster, delete_cluster, set_cluster). The server returns structured JSON responses with data, status, and metadata to support automated workflows.
To use it, configure the MCP client to point to the manticore-search MCP server entry (as shown in the README example). Ensure the Manticore URL is reachable and that the server environment is started with the correct variables. Tools will be available automatically via the MCP protocol once the server is running and responsive to requests.
Typical usage involves performing a search against a specific table/index, creating or inspecting tables, inserting or updating documents, or managing the Manticore cluster. The API is designed for programmatic usage within AI-assisted workflows, so you can compose sequences of operations (e.g., create a table, insert documents, run a search with highlighting, then fetch cluster status) and handle results in your client’s JSON pipeline.
How to install
Prerequisites:
- Go 1.22 or later
- A running Manticore Search instance accessible at the configured URL
Install steps:
- Install Go if not already installed (follow your OS distribution or go.dev).
- Build the MCP server from source:
git clone https://github.com/krajcik/manticore-mcp-server.git
cd manticore-mcp-server
go build -o manticore-mcp-server
- Run the MCP server (example):
./manticore-mcp-server
- Verify the server is up and reachable by the MCP client and by hitting the Manticore URL if needed.
- Configure environment variables or flags as described in the configuration section to customize behavior (MANTICORE_URL, MAX_RESULTS_PER_QUERY, REQUEST_TIMEOUT, DEBUG).
Additional notes
Tips and common considerations:
- Ensure Manticore Search is running and accessible at the configured MANTICORE_URL (default http://localhost:9308).
- Adjust MAX_RESULTS_PER_QUERY to balance performance and result depth; higher values may impact latency.
- Use REQUEST_TIMEOUT to guard long-running queries; tune to your workload.
- The MCP server supports debugging mode via DEBUG=true for verbose logs.
- For complex boolean queries, verify the HTTP API endpoint (e.g., /search) is reachable if using a separate HTTP layer.
- When deploying, ensure the MCP server process has network access to Manticore and that environment variables are properly propagated to the runtime.
Related MCP Servers
trpc-agent-go
trpc-agent-go is a powerful Go framework for building intelligent agent systems using large language models (LLMs) and tools.
station
Station is our open-source runtime that lets teams deploy agents on their own infrastructure with full control.
tiger-cli
Tiger CLI is the command-line interface for Tiger Cloud. It includes an MCP server for helping coding agents write production-level Postgres code.
gopls
MCP server for golang projects development: Expand AI Code Agent ability boundary to have a semantic understanding and determinisic information for golang projects.
kubernetes
A Model Context Protocol (MCP) server for the Kubernetes API.
gcp-cost
💰 An MCP server that enables AI assistants to estimate Google Cloud costs, powered by Cloud Billing Catalog API and built with Genkit for Go