# Mnemoverse — Persistent Memory API for AI Agents

> One memory. Every AI tool. Write once, recall anywhere.

## What is Mnemoverse

Mnemoverse is a persistent memory API. It stores preferences, decisions, lessons, and context — and makes them available across Claude, ChatGPT, Cursor, VS Code, and any tool. One API key, same memories everywhere.

Not a vector database. Memory that learns (Hebbian associations), forgets (consolidation), and improves over time (feedback loop).

## API

Base URL: https://core.mnemoverse.com/api/v1
Auth: X-Api-Key header (or Authorization: Bearer)
Get your free key: https://console.mnemoverse.com

### Endpoints

POST /memory/write    — Store a memory (content, concepts[], domain)
POST /memory/read     — Search memories (query, top_k, domain)
POST /memory/feedback — Rate usefulness (atom_ids[], outcome -1..1)
GET  /memory/stats    — Memory statistics (total atoms, domains, etc.)
POST /memory/write-batch  — Store up to 500 memories at once
POST /memory/read-batch   — Batch query up to 50 queries
POST /memory/consolidate  — Merge similar memories (sleep-like)

### Quick Example (Python)

pip install mnemoverse

from mnemoverse import MnemoClient
client = MnemoClient(api_key="mk_live_YOUR_KEY")
client.write("User prefers Railway for deploys", concepts=["deploy", "railway"])
results = client.read("how does the user deploy?")

### Quick Example (curl)

curl -X POST https://core.mnemoverse.com/api/v1/memory/write \
  -H "X-Api-Key: mk_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"content": "User prefers Railway", "concepts": ["deploy", "railway"]}'

curl -X POST https://core.mnemoverse.com/api/v1/memory/read \
  -H "X-Api-Key: mk_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "how does the user deploy?", "top_k": 5}'

## Setup by Tool

### Claude Code (one command)

claude mcp add mnemoverse \
  -e MNEMOVERSE_API_KEY=mk_live_YOUR_KEY \
  -- npx -y @mnemoverse/mcp-memory-server

### Cursor (.cursor/mcp.json)

{
  "mcpServers": {
    "mnemoverse": {
      "command": "npx",
      "args": ["-y", "@mnemoverse/mcp-memory-server"],
      "env": { "MNEMOVERSE_API_KEY": "mk_live_YOUR_KEY" }
    }
  }
}

### VS Code (.vscode/mcp.json)

{
  "servers": {
    "mnemoverse": {
      "command": "npx",
      "args": ["-y", "@mnemoverse/mcp-memory-server"],
      "env": { "MNEMOVERSE_API_KEY": "mk_live_YOUR_KEY" }
    }
  }
}

### Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json)

{
  "mcpServers": {
    "mnemoverse": {
      "command": "npx",
      "args": ["-y", "@mnemoverse/mcp-memory-server"],
      "env": { "MNEMOVERSE_API_KEY": "mk_live_YOUR_KEY" }
    }
  }
}

### Windsurf (~/.codeium/windsurf/mcp_config.json)

Same format as Cursor config above.

### ChatGPT (Custom GPT with Actions)

Import OpenAPI spec from: https://mnemoverse.com/docs/openapi-gpt.yaml
Auth type: API Key, header name: X-Api-Key, value: mk_live_YOUR_KEY
Manual setup required — see https://mnemoverse.com/docs/api/chatgpt

### Any tool (REST API)

Use the curl examples above. Any tool that can make HTTP requests can use Mnemoverse.

## MCP Server Tools

After setup, your AI gets 4 tools:

memory_write(content, concepts?, domain?)  — Store a memory
memory_read(query, top_k?, domain?)        — Search memories
memory_feedback(atom_ids[], outcome)        — Rate usefulness
memory_stats()                             — Check memory state

## Pricing

Free    $0/mo    1,000 queries/day    10,000 atoms    60 req/min
Pro     $29/mo   50,000 queries/day   500,000 atoms   600 req/min
Team    $149/mo  500,000 queries/day  5,000,000 atoms  3,000 req/min

Enterprise — custom pricing, unlimited, SSO/SAML, dedicated infrastructure.

## Links

Documentation: https://mnemoverse.com/docs/
Console (sign up): https://console.mnemoverse.com
API Reference: https://mnemoverse.com/docs/api/reference
OpenAPI spec (for GPT Actions): https://mnemoverse.com/docs/openapi-gpt.yaml
Python SDK: pip install mnemoverse
MCP Server: npx @mnemoverse/mcp-memory-server
GitHub: https://github.com/mnemoverse