Structured Memory

The Structured Memory MCP Server lets you create and manage living, markdown-based memory documents for focused projects. It keeps your project context organized over time, automatically updating as you converse with your AI assistant, so you can build deep, traversable context for areas like travel planning, research, real estate searches, investments, product planning, or career development.

How to use

You use this MCP server with an MCP client to create, view, and modify structured memory documents. Start by creating a memory document for your focused project, then spin up a project that references that memory. As you chat, your AI assistant will check the available memory and progressively enrich it with new details like destinations, criteria, findings, or priorities. You can also prompt the AI to add or adjust memory sections on demand, or to ask you targeted questions to fill gaps in your knowledge.

Practical usage patterns you’ll likely follow:

• Create a new memory document for your project (for example, a travel plan or research topic).
• Create a new project for conversations in that area and include the usage instructions to guide memory updates.
• Start new chats by leveraging the memory’s current content; your AI will update the memory with relevant context as conversations continue.
• Periodically ask the AI to append or adjust memory sections or prompts to collect missing details (e.g., preferences, constraints, or timelines).

How to install

Choose one of the installation options below to get the Structured Memory MCP Server up and running on your machine.

# Option 1: Install from npm (when published)
npm install -g @nmeierpolys/mcp-structured-memory

If you prefer to install from source, run these steps to clone the repository, install dependencies, and build the project.

git clone https://github.com/nmeierpolys/mcp-structured-memory.git
cd mcp-structured-memory
npm install
npm run build

Available tools

create_memory

Create a new memory document with optional initial content.

list_memories

List all available memory documents.

get_memory_summary

Get a high-level summary of a memory document.

get_section

Retrieve a specific section from a memory document.

get_full_memory

Retrieve the complete content of a memory document.

search_within_memory

Search for information within a memory document.

update_section

Update an entire section of a memory document.

add_to_list

Add an item to a list section.

update_list_item

Update an existing item in a list.

move_list_item

Move an item between list sections.