MCP.so
ログイン

Memra

@usememra

リポジトリ

Memra について

Memra MCP server — persistent memory for AI agents. EU-hosted, privacy-first, sub-100ms hybrid recall.

基本情報

カテゴリ

メモリとナレッジ

トランスポート

stdio

公開者

usememra

投稿者

vonsensey

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "memra": {
      "url": "https://usememra.com/mcp",
      "headers": {
        "Authorization": "Bearer memra_live_YOUR_KEY"
      }
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Memra?

Memra is a remote MCP (Model Context Protocol) server providing persistent memory for AI agents, hosted in Helsinki (EU). It offers privacy-first, deterministic recall in under 100ms with no LLM in the hot path, and has a free tier with no credit card requirement.

How to use Memra?

Obtain an API key at usememra.com, then configure Memra as an MCP server in your client. For Claude Code use claude mcp add --transport http memra https://usememra.com/mcp --header "Authorization: Bearer memra_live_YOUR_KEY". For Cursor, Cline, or any MCP client, add the endpoint and auth header to your mcp.json file.

Key features of Memra

  • memra_remember / memra_recall for storing and retrieving facts with hybrid semantic + lexical search
  • Write-time contradiction detection: storing a conflicting fact returns conflicts[]
  • Read-your-writes with revision tokens and wait_for_revision for deterministic recall
  • memra_supersede and memra_history for full audit trails when memories are retired
  • Staleness signals (staleness_score / last_confirmed) on every result
  • Token-budget recall via max_tokens to fit context windows
  • memra_bootstrap for priority context at session start
  • Multi-tenant namespaces, PII masking (7 EU languages), and read-only scoped keys

Use cases of Memra

  • Give AI agents persistent memory that survives across conversations and sessions
  • Detect and resolve contradictory facts during storage (e.g., outdated user preferences)
  • Retrieve facts exactly matching identifiers, error codes, or semantics within a token budget
  • Manage memory lifecycle with explicit superseding and history for audit purposes
  • Provide priority context at the start of each session via bootstrap

FAQ from Memra

Where is data stored?

Data is hosted in the EU (Helsinki) with a privacy-first approach.

Is there a free tier?

Yes, Memra offers a free tier with no credit card required.

Can Memra run offline?

Yes, an offline alternative called memra-local is available via pip install memra-local that uses SQLite and ONNX embeddings with zero cloud calls.

What transport protocol does Memra use?

Memra uses Streamable HTTP at the endpoint https://usememra.com/mcp.

How is authentication handled?

Authentication is done via a Bearer token sent as an HTTP header: Authorization: Bearer memra_live_YOUR_KEY.

コメント

「メモリとナレッジ」の他のコンテンツ

Zettelkasten MCP Server

entanglr

A Model Context Protocol (MCP) server that implements the Zettelkasten knowledge management methodology, allowing you to create, link, explore and synthesize atomic notes through Claude and other MCP-compatible clients.

159

Memory

modelcontextprotocol

Model Context Protocol Servers

88.1K

Solomd

zhitongblog

A markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.

425

Ultimate Google Docs & Drive MCP Server

a-bonus

The Ultimate Google Docs, Sheets, Drive, Gmail, & Google Calendar MCP Server. This MCP (primarily for use in Claude Desktop) gains full access to your google suite and lets claude do its thing.

605

Anytype MCP Server

anyproto

An MCP server enabling AI assistants to interact with Anytype - your encrypted, local and collaborative wiki - to organize objects, lists, and more through natural language.

467