Frontier-Compute/zcash-mcp
Zcash MCP server for AI agents. 12 tools for shielded operations, memo decoding, ZAP1 attestation. Works with Claude, ChatGPT, OpenClaw.
README
zcash-mcp
Zcash MCP server. Connects AI agents to shielded Zcash operations. First privacy chain on the MCP registry.
MCP (Model Context Protocol) is the standard way for AI models to call external tools. This server exposes 12 Zcash tools that any MCP client can use - Claude Desktop, ChatGPT, OpenClaw, or anything that speaks the protocol.
Tools
| Tool | What it does |
|---|---|
get_balance |
ZAP1 lifecycle events and anchor status for a wallet hash |
send_shielded |
Generate a zcash: payment URI (ZIP 321) |
decode_memo |
Decode shielded memos - ZAP1 typed, ZIP 302, text, binary |
attest_event |
Write a ZAP1 attestation to the Zcash blockchain |
verify_proof |
Verify a ZAP1 Merkle proof |
get_stats |
ZAP1 protocol stats (leaves, anchors, types) |
get_block_height |
Current chain height from Zebra |
lookup_transaction |
Raw transaction data by txid |
get_anchor_history |
All ZAP1 Merkle root anchors with txids and block heights |
get_anchor_status |
Current Merkle tree state: root, unanchored leaves, recommendation |
get_events |
Recent ZAP1 attestation events with type, wallet hash, leaf hash |
get_agent_status |
Attestation summary for a ZAP1 agent ID |
Install
npx @frontiercompute/zcash-mcp
Or install globally:
npm install -g @frontiercompute/zcash-mcp
Configuration
Environment variables:
| Variable | Default | Description |
|---|---|---|
ZEBRA_RPC_URL |
http://127.0.0.1:8232 |
Zebra node JSON-RPC endpoint |
ZAP1_API_URL |
https://pay.frontiercompute.io |
ZAP1 attestation API |
ZAP1_API_KEY |
none | API key for attest_event |
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"zcash": {
"command": "npx",
"args": ["@frontiercompute/zcash-mcp"],
"env": {
"ZEBRA_RPC_URL": "http://127.0.0.1:8232",
"ZAP1_API_KEY": "your-key-here"
}
}
}
}
Any MCP client
The server communicates over stdio using JSON-RPC. Point your MCP client at the zcash-mcp binary.
Build from source
git clone https://github.com/Frontier-Compute/zcash-mcp.git
cd zcash-mcp
npm ci
npm run build
node dist/index.js
Testing
Offline verification covers the built stdio server and a clean-room install from the packed npm tarball:
npm run test:offline
Live verification hits a real Zebra RPC and ZAP1 API:
ZEBRA_RPC_URL=http://127.0.0.1:8232 \
ZAP1_API_URL=http://127.0.0.1:3080 \
ZAP1_API_KEY=your-key-here \
npm run test:live
test:live drives the MCP server over stdio and exercises the live tool surface, not just the underlying HTTP endpoints. Set ZAP1_AGENT_ID if you want the get_agent_status check to target a specific deployed agent.
GitHub Actions mirrors that split:
.github/workflows/offline-ci.ymlruns deterministic packaging and MCP handshake checks on every push and pull request..github/workflows/live-e2e.ymlruns secret-backed live checks onmain, on a schedule, and by manual dispatch.
Dependencies
- A running Zebra node for chain queries (get_block_height, lookup_transaction)
- The ZAP1 API at pay.frontiercompute.io for attestation tools (get_balance, attest_event, verify_proof, get_stats, get_anchor_history, get_anchor_status, get_events, get_agent_status)
- Memo decoding works locally with no external dependencies
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。