Calypso Multimodal RAG MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 121 installs a week from npm. Last commit 28 Aug 2026.
Calypso multimodal RAG for grounded answers from docs, images, charts, and knowledge.
Today is the operative word: we check Calypso Multimodal RAG MCP every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add multimodal-rag-mcp-server -e 'CALYPSO_API_KEY=<CALYPSO_API_KEY>' -- npx -y @calypsohq/multimodal-rag-mcp-server
{
"mcpServers": {
"multimodal-rag-mcp-server": {
"command": "npx",
"args": [
"-y",
"@calypsohq/multimodal-rag-mcp-server"
],
"env": {
"CALYPSO_API_KEY": "<CALYPSO_API_KEY>"
}
}
}
}
[mcp_servers.multimodal-rag-mcp-server]
command = "npx"
args = ["-y", "@calypsohq/multimodal-rag-mcp-server"]
env = { CALYPSO_API_KEY = "<CALYPSO_API_KEY>" }
{
"mcpServers": {
"multimodal-rag-mcp-server": {
"command": "npx",
"args": [
"-y",
"@calypsohq/multimodal-rag-mcp-server"
],
"env": {
"CALYPSO_API_KEY": "<CALYPSO_API_KEY>"
}
}
}
}
{
"servers": {
"multimodal-rag-mcp-server": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@calypsohq/multimodal-rag-mcp-server"
],
"env": {
"CALYPSO_API_KEY": "<CALYPSO_API_KEY>"
}
}
}
}
Put your own value in place of
<CALYPSO_API_KEY> (Calypso API key used to access the Calypso Responses API).
The author’s registry entry marks it as required.
Optional settings from the registry entry, left out of the block above:
CALYPSO_API_BASE_URL (Optional Calypso-compatible API base URL. Defaults to https://api.calypso.so/v1).
Self-hosted memory for AI agents over MCP: knowledge graph, vectors, graph RAG, cited answers.
Local-first personal knowledge base. Cited answers from your vault via vector + graph retrieval.
97-domain knowledge graphs for AI agents — 4× F1 of RAG, 11× fewer tokens. MCP-native.
Ingest and analyze Terraform modules from multiple repositories for AI RAG systems
Lightweight self-hosted RAG knowledge base with hybrid retrieval and cited answers.
Permission-aware RAG for agents — ingest docs, hybrid search, cited answers (hosted API).
Docs + source RAG for Pipe language: semantic search, cited answers, code symbol lookup.
MCP server for NVIDIA NIM - 50+ LLMs, multimodal, image gen, embeddings, reranking
Answers built from our own checks of this server.