retrieve_semantically_similar_internal_knowledgeLABSCO SUMMARY
One search across everything you have connected.
The two search modes do different jobs: the semantic one takes a passage and finds what resembles it, the smart search takes a question and works across mail, chat and documents at once - and what you learn can be written back into the same repository.
What it is
An MCP server over Rememberizer's document and knowledge API. It searches your personal or team knowledge repository two ways: semantic similarity against a block of text, and an agentic search from a plain query that reaches Slack discussions, Gmail, Dropbox, Google Drive and uploaded files. It also lists what is connected and saves new content back.
What you get
- Semantically similar passages for a block of up to 400 words, with an optional date range
- An agentic search across Slack, Gmail, Dropbox, Google Drive and uploaded files
- The knowledge systems currently connected
- The documents in your personal or team repository
- New content saved into the repository
- Your Rememberizer account details
- Documents and Slack discussions available as MCP resources
Requirements
A Rememberizer account with your sources already connected - this searches what Rememberizer has indexed, so the integrations are configured there rather than here. The project describes itself as in development, with functionality subject to change.
Setup effort
One command plus a key — uvx mcp-server-rememberizer, then supply credentials
USE CASES
01
Finding the thread where something was decided
The agentic search reaches Slack and mail together, so you do not have to guess which one it was in.
02
Locating passages that resemble a paragraph you are writing
Send the paragraph itself, up to 400 words, and get the closest chunks back.
03
Saving a decision so it is searchable later
Write it into the repository as it is made, rather than after the fact.
04
Narrowing a search to a period
Start and end timestamps are parameters on the similarity search.
05
Checking what the assistant can actually see
List the connected systems and the documents behind them before you trust an empty result.
6 tools
Listed in project docs. Your client may show a different list.
smart_search_internal_knowledgelist_internal_knowledge_systemsrememberizer_account_informationlist_personal_team_knowledge_documentsremember_thisBEFORE YOU INSTALL
What to know first
Tools you get
6 tools named in the project's documentation.
Requires
- uvx on PATH
- REMEMBERIZER_API_TOKEN in your environment
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
uvx mcp-server-rememberizerMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"rememberizer": {
"command": "uvx",
"args": [
"mcp-server-rememberizer"
],
"env": {
"REMEMBERIZER_API_TOKEN": "<YOUR_REMEMBERIZER_API_TOKEN>"
}
}
}
}
