ExpertPack MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 424 installs a week from pypi. Last commit 20 Oct 2026.
Serve ExpertPack knowledge packs over MCP with hybrid search, graph traversal and provenance.
Today is the operative word: we check ExpertPack 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 ep-mcp -- uvx ep-mcp
{
"mcpServers": {
"ep-mcp": {
"command": "uvx",
"args": [
"ep-mcp"
]
}
}
}
[mcp_servers.ep-mcp]
command = "uvx"
args = ["ep-mcp"]
{
"mcpServers": {
"ep-mcp": {
"command": "uvx",
"args": [
"ep-mcp"
]
}
}
}
{
"servers": {
"ep-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"ep-mcp"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
GEMINI_API_KEY (Gemini API key for embeddings (or set AZURE_OPENAI_API_KEY when using Azure OpenAI)), AZURE_OPENAI_API_KEY (Azure OpenAI key, if the config uses the Azure embedding provider).
Local, explainable knowledge base for AI agents: hybrid search with provenance over SQLite.
Local, explainable knowledge base for AI agents: hybrid search with provenance over SQLite.
ESG knowledge base for AI agents: hybrid search, glossary, frameworks, industries, graph traversal.
ESG knowledge base for AI agents: hybrid search, glossary, frameworks, industries, graph traversal.
ESG knowledge base for AI agents: hybrid search, glossary, frameworks, industries, graph traversal.
Agent-native knowledge OS on Markdown: typed graph, hybrid search, and compiler over MCP.
High-performance knowledge graph MCP server for AI coding agents with semantic search.
Persistent memory for AI tools. Local-first knowledge graph with hybrid search.
Answers built from our own checks of this server.