Bundle
RAG & vector search stack
Build retrieval-augmented apps: four leading vector databases (Qdrant, Pinecone, Weaviate, Chroma) plus AWS Knowledge Bases — compare, prototype, and let your assistant query embeddings directly.
5 servers · created Jun 11, 2026
Install all of them at once
Drop into your claude_desktop_config.json or Cursor MCP settings. Replace the ${…} placeholders with your own values.
{
"mcpServers": {
"aws-kb": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-aws-kb-retrieval"
],
"env": {
"AWS_ACCESS_KEY_ID": "${AWS_KEY}",
"AWS_SECRET_ACCESS_KEY": "${AWS_SECRET}",
"AWS_REGION": "us-east-1"
}
}
}
} Or install one-by-one
uvx mcp-server-qdrant
npm install -g mcp-pinecone
uvx mcp-server-weaviate
uvx chroma-mcp
npm install -g @modelcontextprotocol/server-aws-kb-retrieval Like this stack? Build your own.
Free account, 30 seconds. Save servers, ship public or private bundles, install one-click into Claude Desktop or Cursor.
Servers in this bundle (5)
- Qdrant · Qdrant · Data
Vector search via Qdrant — semantic-memory store the model can write to and query.
- Pinecone · Community · Data
Vector search via Pinecone — managed-cloud RAG store with hybrid search support.
- Weaviate · Community · Data
Vector + structured search via Weaviate — multimodal RAG with built-in modules.
- Chroma · Chroma · Data
Local-first vector database via Chroma — embedded, no separate service needed.
- AWS Knowledge Base · Anthropic · Data
Query a Bedrock Knowledge Base — RAG retrieval over your indexed AWS documents.