MCP.so
ログイン

Assay Protocol Ai Agent Trust MCP Server

@Grandionn

基本情報

カテゴリ

AI とエージェント

トランスポート

stdio

作者

Grandionn

投稿者

Granth

Assay Protocol Ai Agent Trust MCP Server について

Trust infrastructure for AI agents on Base. USDC staking, escrow, reputation scoring, 2000+ ERC-8004 agents indexed.

Assay Protocol Ai Agent Trust MCP Server の設定

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": {
    "assay": {
      "command": "npx",
      "args": [
        "-y",
        "@assaylabs/mcp-server"
      ]
    }
  }
}

Assay Protocol Ai Agent Trust MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is Assay Protocol Ai Agent Trust?

Assay Protocol Ai Agent Trust is an MCP server that exposes the Assay Discovery and trust infrastructure tools. It enables AI agents and tools like Claude Desktop and Cursor to query agent trust assessments, perform semantic search over indexed agents, and retrieve ecosystem statistics via the standard MCP stdio transport.

How to use Assay Protocol Ai Agent Trust?

Install the package globally or run it directly with npx @assaylabs/mcp-server. Set the ASSAY_API_URL environment variable (defaults to https://assay-discovery-api.onrender.com). Add the server configuration to your Claude Desktop or Cursor MCP config file, specifying the command and environment variable.

Key features of Assay Protocol Ai Agent Trust

  • Agent trust lookup by ERC-8004 address
  • Semantic search over indexed agents
  • Optional minimum trust score filtering
  • Top-5 ranked search results from Assay Discovery
  • Ecosystem statistics (total agents, average trust, etc.)
  • Works with Claude Desktop and Cursor MCP clients

Use cases of Assay Protocol Ai Agent Trust

  • Verify the trust score of a specific agent before integration
  • Find agents by natural language queries like "smart contract security"
  • Monitor the health and distribution of trust in the ecosystem
  • Filter agents by minimum trust threshold for safe selections
  • Get a quick overview of total indexed agents and top performers

よくある質問

1

What transport does the server use?

The server uses standard MCP stdio transport.

2

What environment variable must be set?

You must set `ASSAY_API_URL` (or fallback `VITE_API_URL`). The default is `https://assay-discovery-api.onrender.com`.

3

What tools does the server expose?

Three tools: `check_agent_trust` (by address), `search_agents` (by query and optional minScore), and `get_ecosystem_stats` (no input).

4

How do I configure it for Claude Desktop?

Add a JSON entry to your MCP config with `command: "npx"`, `args: ["-y", "@assaylabs/mcp-server"]`, and the `ASSAY_API_URL` env var.

5

What is the package name and license?

The package is `@assaylabs/mcp-server` and is licensed under MIT.

コメント

「AI とエージェント」の他のコンテンツ