Pikelet runs on your own machine — the client starts it, so there is no endpoint to ping. 144 installs a week from npm. Last commit 18 Sep 2026.
Search self-contained .pikelet knowledge artifacts over MCP — no vector DB, no embedding API.
Today is the operative word: we check Pikelet 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 pikelet -- npx -y pikelet
{
"mcpServers": {
"pikelet": {
"args": [
"-y",
"pikelet"
],
"command": "npx"
}
}
}
[mcp_servers.pikelet]
command = "npx"
args = ["-y", "pikelet"]
{
"mcpServers": {
"pikelet": {
"args": [
"-y",
"pikelet"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"pikelet": {
"args": [
"-y",
"pikelet"
],
"command": "npx"
}
}
}
Minimal RAG-over-a-corpus MCP retrieval: search_knowledge returns cited chunks. Local embeddings.
MCP memory server for AI coding assistants. Windows Service, ONNX embeddings, SQLite vector search.
Persistent local AI memory. SQLite + FTS5 + vector search, knowledge graph. No cloud, no API keys.
AI-powered vector knowledge graph MCP server for agents. Supports add node, add edge, semantic node
MCP server for the xAI Grok API — chat, vision, search, and embeddings
Persistent memory and knowledge graph for AI assistants — keyword + vector + graph search.
MCP server for Obsidian Smart Connections. Semantic search using your vault's embeddings.
Self-hosted MCP memory server: a typed knowledge graph on Postgres 16 and pgvector.
Answers built from our own checks of this server.