get_templateLABSCO SUMMARY
Say it once instead of at the top of every chat.
It keeps one copy of the context you would otherwise retype — how you want answers, what stack you are on, who you are — and keeps that copy identical across the Claude apps you move between.
What it is
A local-first server that keeps one set of personal context — preferences, personal details, code standards — and synchronises it across Claude Desktop, Claude Code, Windsurf and Claude on the web. The data stays on your own machine.
What you get
- One memory, updated in natural language, that appears in every Claude interface you work in
- Separate sections for preferences, personal details and code standards, so the same rules follow you between projects
- Local storage — the data stays on your own device rather than on a service
- A caching layer aimed at making repeated memory reads cheap
- Runs over stdio with
npm start, or over HTTP withnpm run start:http
Requirements
Node with npm: clone the repository, npm install, npm run build, then npm start. No account and no key — the memory lives in a file on your machine. Also installable through Smithery.
Setup effort
Build from source — git clone https://github.com/Jktfe/myaimemory-mcp.git
USE CASES
01
Stopping the repeat-yourself opening
The preferences you would otherwise paste at the start of a session are already in place when it begins.
02
Keeping code standards consistent across projects
Standards live in the shared memory rather than in one project's instructions, so a new repository starts with them.
03
Changing a preference in one place
Update it once and the other interfaces get the same text, instead of three copies drifting apart.
9 tools
Listed in project docs. Your client may show a different list.
get_sectionupdate_sectionupdate_templatelist_presetsload_presetcreate_presetsync_platformslist_platformsBEFORE YOU INSTALL
What to know first
Tools you get
9 tools named in the project's documentation.
Requires
node on PATH
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
# Clone repository
git clone https://github.com/Jktfe/myaimemory-mcp.git
cd myaimemory-mcp
# Install dependencies
npm install
# Build TypeScript code
npm run build
# Start MCP server (with stdio transport)
npm start
# Or start with HTTP transport
npm run start:httpMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"myai-memory-sync": {
"command": "node",
"args": [
"<YOUR_PATH_TO_MYAIMEMORY_MCP>/dist/unified-cli.js",
"server"
],
"env": {
"TEMPLATE_PATH": "<YOUR_PATH_TO_CUSTOM>/template.md",
"ENABLE_ANTHROPIC": "true",
"ANTHROPIC_API_KEY": "<YOUR_ANTHROPIC_API_KEY>"
}
}
}
}Windsurf
WINDSURF · JSON
{
"mcpServers": {
"myai-memory-sync": {
"command": "node",
"args": [
"<YOUR_PATH_TO_MYAIMEMORY_MCP>/dist/unified-cli.js",
"server"
]
}
}
}
