What is visual memory and how does the knowledge graph work?

Visual memory stores conversation context and structured data in a knowledge graph format. The server persists this data to a memory.json file (configured via MEMORY_FILE_PATH) so context survives across sessions and can be queried for relationships between stored concepts.

How do I set up the server and configure file access?

Install via `npx visual-memory-context-server@1.0.1`. Set MEMORY_FILE_PATH to your desired storage location and ALLOWED_DIRECTORIES to a comma-separated list (or JSON array) of directories the server can read/write. Only these directories will be accessible.

Can I use this server to persist state across multiple conversations?

Yes. The knowledge graph is stored in memory.json, so all entities, relationships, and context are retained between sessions. You can query and update the graph in new conversations using the stored data.

What's the difference between this and a simple JSON file storage?

This server adds knowledge graph capabilities—structured relationships and entity connections—rather than flat data storage. You can query connections between concepts, not just retrieve raw values, making it better for complex context management.

What happens if ALLOWED_DIRECTORIES is not set or is empty?

The server will restrict file access based on your configuration. If not set, refer to the documentation for default behavior. Always explicitly configure ALLOWED_DIRECTORIES to ensure the server can access the directories where you store memory.json.

MCP ServerSTDIOOfficialv1.0.1

Gabrielahdzmicrosoft MCP Server

Store and retrieve visual memory with knowledge graph capabilities for AI context. Developers use this to persist multi-turn conversation state and structured knowledge across sessions.

io.github.GabrielaHdzMicrosoft/mcp-server

Try MCP with AI models View source

Hosted URL

Local install

Transport

STDIO

Auth

No auth required

What the Gabrielahdzmicrosoft MCP server does

How models use it and what it is built for.

Store and retrieve visual memory with knowledge graph capabilities for AI context. Developers use this to persist multi-turn conversation state and structured knowledge across sessions.

Connect to Gabrielahdzmicrosoft

Local install — runs as a subprocess.

npx visual-memory-context-server@1.0.1

Environment variables

Configuration this server reads at startup.

MEMORY_FILE_PATH
Path to the memory.json file for knowledge graph storage
ALLOWED_DIRECTORIES
Comma-separated list of directories the server can access, or JSON array format

Resources

Where to find authoritative docs and source for Gabrielahdzmicrosoft.

Source repository

Example prompts for Gabrielahdzmicrosoft

Paste any of these into Agent Studio after connecting Gabrielahdzmicrosoft.

Add a new entity to the knowledge graph with relationships to existing nodes
Retrieve all stored memories related to user preferences and conversation history
Query the knowledge graph to find connections between stored concepts
What directories can this server access and how do I configure ALLOWED_DIRECTORIES?

Gabrielahdzmicrosoft MCP server — FAQ

Common questions about connecting and running Gabrielahdzmicrosoft.

What is visual memory and how does the knowledge graph work?
Visual memory stores conversation context and structured data in a knowledge graph format. The server persists this data to a memory.json file (configured via MEMORY_FILE_PATH) so context survives across sessions and can be queried for relationships between stored concepts.
How do I set up the server and configure file access?
Install via `npx visual-memory-context-server@1.0.1`. Set MEMORY_FILE_PATH to your desired storage location and ALLOWED_DIRECTORIES to a comma-separated list (or JSON array) of directories the server can read/write. Only these directories will be accessible.
Can I use this server to persist state across multiple conversations?
Yes. The knowledge graph is stored in memory.json, so all entities, relationships, and context are retained between sessions. You can query and update the graph in new conversations using the stored data.
What's the difference between this and a simple JSON file storage?
This server adds knowledge graph capabilities—structured relationships and entity connections—rather than flat data storage. You can query connections between concepts, not just retrieve raw values, making it better for complex context management.
What happens if ALLOWED_DIRECTORIES is not set or is empty?
The server will restrict file access based on your configuration. If not set, refer to the documentation for default behavior. Always explicitly configure ALLOWED_DIRECTORIES to ensure the server can access the directories where you store memory.json.

Skip the local setup — run MCP in your browser

MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.

Open Agent Studio