MCP.so
ログイン
T

Thbttb

@qu3ai

Thbttb について

概要はまだありません

基本情報

カテゴリ

その他

トランスポート

stdio

公開者

qu3ai

投稿者

Dharma Reddy

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": {
    "github": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "GITHUB_PERSONAL_ACCESS_TOKEN",
        "ghcr.io/github/github-mcp-server"
      ],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

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 QU3?

QU3 is a command-line client application that establishes quantum-safe secure communication with Multi-Compute Provider (MCP) servers. It uses NIST post‑quantum cryptography (Kyber KEM and SPHINCS+ signatures) to authenticate clients, encrypt payloads, and verify server attestations. Developers and operators who need post‑quantum security for MCP workloads can use QU3 to run inference, execute agent workflows, and update policies.

How to use QU3?

Install dependencies, configure config.yaml (optional), start the mock server (python -m scripts.mock_mcp_server), then run client commands from another terminal: generate keys with python -m src.main generate-keys, send a single inference with run-inference <model> <json_data>, chain models with run-agent "<modelA> -> <modelB>" <data>, or update a policy with update-policy --policy-file <file>. The client automatically fetches server public keys if not found locally.

Key features of QU3

  • Uses NIST PQC finalists Kyber-768 for key encapsulation and SPHINCS+-SHA2-128f-simple for signatures
  • Establishes secure sessions via a KEM handshake endpoint (/kem-handshake/initiate)
  • Encrypts all request/response payloads with AES-256-GCM derived via HKDF-SHA256
  • Provides CLI commands: generate-keys, run-inference, run-agent, update-policy
  • Includes a FastAPI mock server with endpoints for key exchange, inference, and policy updates
  • Comes with unit tests for cryptographic utilities, configuration, and client communication

Use cases of QU3

  • Run inference on a quantum‑safe MCP server while ensuring end‑to‑end encryption and authentication
  • Chain multiple models in a secure agent workflow, passing outputs as inputs with step‑by‑step reporting
  • Send encrypted and signed policy updates to a server to enforce access control rules
  • Test and develop PQC‑enabled MCP integrations using the included mock server

FAQ from QU3

What PQC

コメント

「その他」の他のコンテンツ