Search Memory runs on your own machine — the client starts it, so there is no endpoint to ping. 471 installs a week from npm. Last commit 12 Oct 2026.
Free web search and long-term memory for AI assistants. No API keys; data stays local.
We read the source, 7 h ago · rules ccca72095570
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
postinstall: playwright install chromium || node -e "process.exit(0)"
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
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 search-memory-mcp -- npx -y search-memory-mcp
{
"mcpServers": {
"search-memory-mcp": {
"args": [
"-y",
"search-memory-mcp"
],
"command": "npx"
}
}
}
[mcp_servers.search-memory-mcp]
command = "npx"
args = ["-y", "search-memory-mcp"]
{
"mcpServers": {
"search-memory-mcp": {
"args": [
"-y",
"search-memory-mcp"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"search-memory-mcp": {
"args": [
"-y",
"search-memory-mcp"
],
"command": "npx"
}
}
}
This one needs environment variables set before it will start:
SEARCH_PROVIDERS (Comma-separated search provider order, for example duckduckgo,marginalia,bing), CACHE_DB_PATH (Path of the local SQLite database for cache, memory, and index), INGEST_ALLOWED_DIRS (Comma-separated directories whose files ingest_document may read; local file reading is off when unset), SEARXNG_BASE_URL (Base URL of a SearXNG instance with JSON output, used by the searxng provider), BRAVE_SEARCH_API_KEY (Optional Brave Search API key; the brave provider then uses the official API instead of scraping), READER_FALLBACK_URL (Optional reader service (for example https://markdown.new/) used when a page cannot be fetched directly; sends the page URL to that service).
The author declared them in the registry entry; get the values from the project itself.
AI assistant with RAG, web search, filesystem and memory. MCP sub-agent for Claude Code.
MyWebSearch - multi-engine web search MCP server, CLI and local daemon. No API keys required.
No-API-key web search, page fetching and document reading across many engines. Local-first.
42 local SEO tools for AI assistants. SERP tracking, reviews, keywords, audits, and more.
Keyless multi-engine web search, fetch, and clean-Markdown read for AI agents. No API keys.
Web search, clean page reading & one-call research dossiers for AI agents. No API key.
Web search and web fetch with no API keys. Reads the sites that refuse AI agents.
MCP server providing Google Search, web scraping, and multi-source research tools for AI assistants
Answers built from our own checks of this server.