Gnosem
One memory store that Claude, ChatGPT, Cursor and other MCP clients all read and write
The repository was last updated 24 days ago and is published under MIT.
- Pricing
- Freemium
- Platforms
- API, CLI
- Runs
- Remotely
- GitHub stars
- 0
- Last commit
- 24 days ago
- Language
- JavaScript
About Gnosem
Gnosem is a freemium MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.
The pitch is neutrality between vendors: a fact saved from one assistant can be recalled from any other that uses the same API key. Five tools handle it: memory_write, memory_search, memory_list, memory_forget for soft deletes, and memory_supersede to replace a stale memory with a corrected one.
Memories over 400 characters are compressed on write into a structured-facts form such as TOPIC, PROJECT, DECISION, STACK and PROBLEM using a Workers AI model, while the raw text is kept and returned as content_raw. Compression is skipped if the result is not shorter, and a failed AI call still saves the memory. Each row records provenance in a written_by field and stores a 768-dimension semantic embedding.
The hosted service runs on Cloudflare Workers, D1, Vectorize and Workers AI. An npx gnosem-install command configures Claude Desktop, Claude Code, Cursor, Windsurf and Zed with your key. The repository is MIT licensed.
Install
Run in a terminal
claude mcp add --transport http gnosem https://gnosem.dev/mcpAdd to Settings → Connectors → Add custom connector
https://gnosem.dev/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"gnosem": {
"url": "https://gnosem.dev/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"gnosem": {
"type": "http",
"url": "https://gnosem.dev/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"gnosem": {
"serverUrl": "https://gnosem.dev/mcp"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"gnosem": {
"type": "http",
"url": "https://gnosem.dev/mcp"
}
}
}Tools this server provides (5)
Taken from the project documentation; the running server may differ.
- memory_write
- Save a fact, preference, decision or note.
- memory_search
- Semantic search across your memories.
- memory_list
- List recent memories, newest first.
- memory_forget
- Soft-delete a memory by id.
- memory_supersede
- Replace a stale memory with a corrected one.
What it does
- Cross-vendor memory
- One store shared by any client that speaks MCP or HTTP tools.
- Semantic search
- Finds memories by meaning using embeddings.
- Auto compression
- Long memories are stored as compact structured facts alongside the raw text.
- Provenance field
- Records which model or client wrote each memory.
- Supersede and forget
- Replace stale facts or soft-delete them.
- One-command install
- npx gnosem-install configures supported editors.
Good for
- Carry preferences from Claude into Cursor
- Record a technical decision once and recall it anywhere
- Correct an outdated fact instead of leaving it conflicting
- Keep context across several AI subscriptions
Pricing
Freemium, from $9 a month
Free: 200 memories; Pro $9/mo or $90/yr
Prices change. Check gnosem.dev before you buy.
Gnosem FAQ
What does it cost?
Free is 200 memories and one API key; Pro is $9 a month or $90 a year with unlimited memories and keys and 1 GB storage.
How do I get a key?
POST your email to /signup, which returns a one-time API key.
Which clients can install automatically?
Claude Desktop, Claude Code, Cursor, Windsurf and Zed.
Is the raw text kept?
Yes. Raw prose is preserved and returned as content_raw.
Is Gnosem free?
Gnosem has a free plan, with paid plans for more usage or features. Paid plans start at $9 a month.
How do I add Gnosem to Claude Code?
Run this command in your terminal: claude mcp add --transport http gnosem https://gnosem.dev/mcp
What are the best alternatives to Gnosem?
The closest alternatives to Gnosem are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.
Alternatives to Gnosem

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