Dawnmere
Shared local memory for Claude Code, Codex and Cursor on a Mac
- Pricing
- Free
- Platforms
- macOS, CLI
- Runs
- Locally
- Version
- 0.1.1
- Listing updated
- 3 days ago
About Dawnmere
Dawnmere is a free MCP server for knowledge and memory. It runs locally on your machine and is started by your AI client.
The connector is a small bridge: each AI app launches it, and it passes recall, remember and forget calls to the Dawnmere app running in the background on the same Mac. The connector stores nothing and sends nothing over the network. Without Dawnmere running, every call answers that Dawnmere is not running.
Memory lives in one SQLite file with local search, with no account and no embedding service. Each app has an access level, and the name you set in LFLABS_MEMORY_CLIENT decides what that app may see. An app that only names itself, including Claude Code and Claude Desktop, sees work notes only, while Cursor and Codex see more once widened with the dawnmere access command.
It is an early beta that needs Apple silicon, macOS 14 or later and Node.js 18 or later.
Install
Run in a terminal
claude mcp add dawnmere-connector -- npx -y dawnmere-connectorAdd to claude_desktop_config.json
{
"mcpServers": {
"dawnmere-connector": {
"command": "npx",
"args": [
"-y",
"dawnmere-connector"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"dawnmere-connector": {
"command": "npx",
"args": [
"-y",
"dawnmere-connector"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"dawnmere-connector": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"dawnmere-connector"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"dawnmere-connector": {
"command": "npx",
"args": [
"-y",
"dawnmere-connector"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"dawnmere-connector": {
"command": "npx",
"args": [
"-y",
"dawnmere-connector"
]
}
}
}Configuration
LFLABS_MEMORY_CLIENTA name for this app (cursor, codex, windsurf...). Dawnmere shows it and uses it for what this app may see.
Tools this server provides (4)
Taken from the project documentation; the running server may differ.
- recall
- Retrieve stored memories.
- remember
- Save a new memory.
- forget
- Remove a stored memory.
- access_status
- Show what the calling app may see.
What it does
- One memory, many apps
- Claude, Cursor, Codex and others reach the same store.
- Local only
- A single SQLite file on your Mac, never uploaded.
- Per-app access levels
- Each app sees only what its configured name allows.
- Source tracking
- Every memory records where it came from.
- Secret refusal
- API keys and tokens are refused.
Good for
- Tell Claude Code a decision and have Cursor recall it later
- Keep project conventions across several AI apps
- Limit what each AI app can read
- Wipe a remembered item with forget
Pricing
Free
Free, no account
Prices change. Check dawnmere.app before you buy.
Dawnmere FAQ
Does it need an account?
No account is needed, and the memory is never uploaded to LF Labs.
What does it need to run?
A Mac with Apple silicon, macOS 14 or later and Node.js 18 or later.
What do I set per app?
LFLABS_MEMORY_CLIENT, such as cursor, windsurf or gemini.
Is it stable?
It is labelled an early beta and Mac only.
Is Dawnmere free?
Yes. Dawnmere is free to use.
How do I add Dawnmere to Claude Code?
Run this command in your terminal: claude mcp add dawnmere-connector -- npx -y dawnmere-connector
What are the best alternatives to Dawnmere?
The closest alternatives to Dawnmere are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.
Alternatives to Dawnmere

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Claude Code Ultimate Guide
MCP serverMCP server for searching the Claude Code Ultimate Guide from any MCP client
Runs locally- Open source
- 6.5K GitHub stars
- AI models and LLM tooling

Nautilus Compass Memory
MCP serverLocal-first long-term memory layer for AI agents, served over MCP
Not checked yet- 5 tools
- Remote
- Open source
- 1.3K GitHub stars
- Knowledge and memory

note1
MCP serverMCP server giving AI assistants search and summaries across your note1 meetings
- 4 tools
- Remote
- Free trial
- 1 GitHub star
- Productivity
MedXpert 图书馆连接器(通用版)
MCP serverLocal MCP server that searches and answers questions over your own Markdown wiki
Runs locally- 4 tools
- Open source
- Knowledge and memory