What does Snap Protocol do that a simple token transfer cannot?
It makes the deposit and withdrawal unlinkable: observers see only shielded pool activity, not the connection between the sending and receiving agents.
@agentzeny
基本情報
概要はまだありません
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": {
"snap": {
"command": "npx",
"args": [
"tsx",
"snap-mcp-server.ts"
],
"env": {
"SNAP_RPC_URL": "https://api.mainnet-beta.solana.com",
"SNAP_POOL_ADDRESS": "B8SyffZKt8LABKogWjH9rZcjY5PV2hyYRCbTxxbcrpFf"
}
}
}
}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。
Snap Protocol is an MCP server that enables private agent-to-agent payments on Solana mainnet using zero-knowledge proofs. It lets two agents conduct unlinkable transfers through shielded pools — one agent deposits tokens and generates a secret note, the other agent withdraws using a Groth16 proof.
Install the SDK via npm install snap-solana-sdk, then use the provided tools: snap_list_pools, snap_deposit, snap_withdraw, and snap_estimate_fee. A typical flow is deposit into a pool, share the resulting secret note off‑chain, and have the receiving agent withdraw with a zero‑knowledge proof.
It makes the deposit and withdrawal unlinkable: observers see only shielded pool activity, not the connection between the sending and receiving agents.
Solana mainnet-beta and the Node.js package `snap-solana-sdk`. No additional infrastructure is needed; the server handles proof generation and submission.
Secret notes are generated during deposit and must be sent off‑chain by the depositor to the intended recipient. They are not stored on-chain.
The protocol currently offers three shielded pools (0.1 SOL, 1 USDC, 10 USDC) and imposes a 0.25% protocol fee on withdrawals. The withdrawal can be direct or via a relayer.
The README does not specify transport or authentication details. The server communicates with the Solana network and exposes the listed tools for MCP clients.
Hindsight: Agent Memory That Learns
Painless E2E Automation for Mobile and Web
PickFu MCP is a Model Context Protocol (MCP) server that connects AI assistants, developers, marketers, product managers, and AI agents to PickFu's consumer research platform. It enables users to create surveys, conduct

DeepMark helps teachers deliver rapid, consistent marking with meaningful feedback for every student — in a fraction of the time. What once took a week, now takes one free period.

Health by Dr Brad works out a health plan from your own results: which screening is due at your age, and evidence-based suggestions, each with the reason and the published guideline or study behind it. Your record is one
コメント