Rememberizer

This MCP server enables you to access and query your Rememberizer internal knowledge, including documents and Slack discussions, through a compact set of tools. It lets you semantically search memory chunks and retrieve relevant context to support your work, learning, and decision making.

How to use

You interact with the Rememberizer MCP server through your MCP client by invoking its tools to search and retrieve internal knowledge. Use retrieve_semantically_similar_internal_knowledge to find semantically similar chunks from your memory repository by providing a block of text. Use smart_search_internal_knowledge to run an agentic search across your personal or team knowledge sources, including Slack discussions, Gmail, and Drive/uploaded files. You can list available knowledge systems with list_internal_knowledge_systems, fetch account details with rememberizer_account_information, and manage documents with list_personal_team_knowledge_documents. To memory a new piece of information for future retrieval, use remember_this with a descriptive name and the content to memorize.

How to install

# Installation and runtime are configured via your client/app as shown below.
# The Rememberizer MCP server is run locally via the stdio interface using uvx.
# No additional setup is required beyond adding the MCP config to your client.

Additional setup and configuration

You can configure your Claude Desktop client (or SkyDeck AI Helper App) to connect to the Rememberizer MCP server by adding the following MCP configuration snippet. This creates a local stdio-based MCP connection that runs the Rememberizer MCP server command.

{
  "mcpServers": {
    "rememberizer": {
      "command": "uvx",
      "args": ["rememberizer-mcp-mcp"]
    }
  }
}

Notes and tips

• The server exposes tools to search and recall internal knowledge. Use clear, contextual queries to improve relevance. - The remember_this tool helps you build a personal or team memory that can be queried later with the other search tools. - If you need to understand what the Rememberizer MCP server can access, consider listing your internal knowledge sources with list_internal_knowledge_systems and checking account details with rememberizer_account_information.

Troubleshooting

If you encounter issues connecting to the Rememberizer MCP server, verify that your MCP client includes the rememberizer stdio configuration with the exact command uvx and arguments rememberizer-mcp-mcp. Check that the client’s claude_desktop_config.json (or equivalent) contains the proper mcpServers entry and that your environment can execute the uvx runtime. If you see authentication or access errors, review any required API keys or environment placeholders your setup might expect and supply valid values.

Available tools

retrieve_semantically_similar_internal_knowledge

Accepts a block of text and returns semantically similar knowledge chunks from your Rememberizer memory repository. Input includes a 400-word limit for the match text, an optional number of results, and optional date filters.

smart_search_internal_knowledge

Performs agentic search across personal/team internal knowledge sources, returning relevant documents from Slack, Gmail, Dropbox, Google Drive, and uploaded files based on a query and optional context.

list_internal_knowledge_systems

Lists available internal knowledge integrations such as Slack discussions, Gmail, Dropbox documents, Google Drive documents, and uploaded files.

rememberizer_account_information

Provides account details for your Rememberizer personal/team knowledge repository, including holder name and email.

list_personal_team_knowledge_documents

Retrieves a paginated list of all documents in your internal knowledge system with options for page and page size.

remember_this

Stores a piece of text in Rememberizer memory so it can be recalled later by the search tools.