mcpField
S
smara

smara

Local
github.com/smara-ioMemory & Knowledge0Updated 29w ago
Repository

About smara

smara is an MCP server that connects AI apps to the Smara Memory API for persistent user memory. Its temporal scoring ranks memories by recency, importance, and access patterns, and it detects and resolves contradictions.

Key features

  1. 01Store user facts with importance scoring
  2. 02Search memories with temporal ranking
  3. 03Generate formatted user context
  4. 04Delete individual memories
  5. 05Check usage and memory counts

Use cases

  1. 01Retain user facts across AI sessions
  2. 02Retrieve relevant memories for an LLM prompt
  3. 03Review memory usage and remove outdated entries

MCP support

The MCP server exposes Smara Memory API tools for storing, searching, retrieving context, deleting memories, and checking usage. It requires an SMARA_API_KEY and can connect through the configured API URL.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "parallelromb-smara": {
      "command": "npx",
      "args": [
        "-y",
        "@smara/mcp-server"
      ],
      "env": {
        "SMARA_API_KEY": "<SMARA_API_KEY>"
      }
    }
  }
}
Run commandnpx -y @smara/mcp-server
Environment variables to set
SMARA_API_KEYSecretYour Smara API key from smara.io

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@smara/mcp-server2.0.2

Tools

5

Saved tool list · updated 2w ago

ParameterTypeRequiredDescription
factstringYesThe fact or preference to remember
user_idstringYesUnique identifier for the user
namespacestringNoMemory namespace for isolation (default: from env or 'default')
importancenumberNoImportance score (0-1). Higher = slower decay.

Details

From the project README

@smara/mcp-server

MCP server for the Smara Memory API — give any AI app persistent memory with Temporal Memory Scoring™.

Quick Start

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "smara": {
      "command": "npx",
      "args": ["-y", "@smara/mcp-server"],
      "env": {
        "SMARA_API_KEY": "your-api-key-here"
      }
    }
  }
}

Get a free API key at smara.io (10,000 memories, no credit card).

Tools

ToolDescription
store_memoryStore a fact about a user with importance scoring
search_memoriesSemantic search with Temporal Memory Scoring™
get_user_contextPre-formatted context string for LLM system prompts
delete_memoryDelete a specific memory
get_usageCheck plan limits and memory count

How It Works

Smara uses Temporal Memory Scoring™ — a proprietary ranking system that makes AI memory work like human recall. Memories naturally fade over time, modulated by importance and access patterns. Recent, critical memories surface first. Stale, trivial ones fade. Contradictions are auto-detected and resolved.

Works With

  • Claude Desktop
  • Claude Code
  • Cursor
  • Windsurf
  • VS Code (GitHub Copilot)
  • Any MCP-compatible client

Environment Variables

VariableRequiredDefault
SMARA_API_KEYYes—
SMARA_API_URLNohttps://api.smara.io

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.