Golden Dataset MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 77 installs a week from pypi. Last commit 21 Jun 2026.
Version-controlled golden datasets and RAG evaluation, no API key needed.
Today is the operative word: we check Golden Dataset 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 golden-dataset-mcp -- uvx golden-dataset-mcp
{
"mcpServers": {
"golden-dataset-mcp": {
"args": [
"golden-dataset-mcp"
],
"command": "uvx"
}
}
}
[mcp_servers.golden-dataset-mcp]
command = "uvx"
args = ["golden-dataset-mcp"]
{
"mcpServers": {
"golden-dataset-mcp": {
"args": [
"golden-dataset-mcp"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"golden-dataset-mcp": {
"args": [
"golden-dataset-mcp"
],
"command": "uvx"
}
}
}
Generate QA datasets & evaluate RAG systems with failure diagnosis. Any LLM.
Nigerian & Commonwealth legal reasoning, RAG search, contract analysis, and citation verification.
Text extraction, keyword extraction, language detection, and chunking for RAG
Create and manage vector database indexes, backups, and collections
Search self-contained .pikelet knowledge artifacts over MCP — no vector DB, no embedding API.
Inject, parse, and strip [N] citation markers in RAG outputs.
Trace, evaluate, and optimize your LLM, RAG, and agent apps with Noveum observability.
MCP RAG server with hybrid search, multi-KB support, and AI-powered chunk contextualization.
Answers built from our own checks of this server.