Opensolr MCP server
{
"mcpServers": {
"opensolr": {
"command": "uvx",
"args": [
"opensolr-mcp"
],
"env": {
"OPENSOLR_API_KEY": "<opensolr_api_key>",
"YOUR_OPENSOLR_API_KEY": "<your_opensolr_api_key>"
}
}
}
}{
"mcpServers": {
"opensolr": {
"command": "uvx",
"args": [
"opensolr-mcp"
],
"env": {
"OPENSOLR_API_KEY": "<opensolr_api_key>",
"YOUR_OPENSOLR_API_KEY": "<your_opensolr_api_key>"
}
}
}
}{
"mcpServers": {
"opensolr": {
"command": "uvx",
"args": [
"opensolr-mcp"
],
"env": {
"OPENSOLR_API_KEY": "<opensolr_api_key>",
"YOUR_OPENSOLR_API_KEY": "<your_opensolr_api_key>"
}
}
}
}{
"servers": {
"opensolr": {
"command": "uvx",
"args": [
"opensolr-mcp"
],
"env": {
"OPENSOLR_API_KEY": "${input:opensolr_api_key}",
"YOUR_OPENSOLR_API_KEY": "${input:your_opensolr_api_key}"
}
}
},
"inputs": [
{
"id": "opensolr_api_key",
"type": "promptString",
"description": "Opensolr api key",
"password": true
},
{
"id": "your_opensolr_api_key",
"type": "promptString",
"description": "Your opensolr api key",
"password": true
}
]
}{
"mcpServers": {
"opensolr": {
"command": "uvx",
"args": [
"opensolr-mcp"
],
"env": {
"OPENSOLR_API_KEY": "<opensolr_api_key>",
"YOUR_OPENSOLR_API_KEY": "<your_opensolr_api_key>"
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall uvx opensolr-mcpversion 0.5.0
What it does
MCP server for Opensolr, a managed Apache Solr service. Exposes hybrid (BM25 + semantic) search, server-side GPU embeddings, document indexing, and grounded RAG answers. For AI agents that need to search and retrieve information from custom indexes.
- Search a knowledge base using both keyword matching and semantic similarity in a single query, with optional filters.
- Add documents to an index with automatic server-side embedding, then retrieve them with grounded AI-generated answers citing the matching documents.
- Index FAQ or documentation content and let agents answer user questions by finding relevant documents and generating responses based on them.
Tools this server exposes
| Tool | Does |
|---|---|
| opensolr_search | Perform hybrid (keyword + semantic) or pure semantic search with optional Solr filters. |
| opensolr_ai_answer | Generate a grounded RAG answer by using top search results as context for the LLM. |
| opensolr_add_documents | Index plain text documents with metadata; embeddings are computed server-side. |
| opensolr_delete_documents | Remove documents from an index by their ID. |
| opensolr_list_indexes | List all indexes in the account. |
| opensolr_index_info | Get detailed information about a specific index. |
| opensolr_create_index | Create a new vector-enabled index in a specified region (us, de, fi). |
| opensolr_vector_regions | Retrieve the live list of available vector-enabled regions. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.