XRPL Data MCP
Integrates multiple XRPL data sources including LOS, Validator History Service, XRPL JSON-RPC, and XRPLMeta to provide comprehensive querying of XRPL network data, accounts, transactions, tokens, validators, and more via MCP tools.
README
XRPL MCP Server
MCP server for:
- LOS (
https://los.prod.ripplex.io) - Validator History Service (
https://data.xrpl.org) - XRPL JSON-RPC (
https://s1.ripple.com:51234) - XRPLMeta (
https://s1.xrplmeta.org)
Setup
npm install
npm start
npm start runs the server over Streamable HTTP at http://127.0.0.1:3000/mcp by default.
For stdio mode (used by some local MCP clients and the all-tools harness), run:
npm run start:stdio
Environment Variables
LOS_BASE_URL(default:https://los.prod.ripplex.io)DATA_XRPL_BASE_URL(default:https://data.xrpl.org)XRPL_RPC_URL(default:https://s1.ripple.com:51234)XRPLMETA_BASE_URL(default:https://s1.xrplmeta.org)MCP_TRANSPORT(default:http, options:httporstdio)MCP_HTTP_HOST(default:0.0.0.0)MCP_HTTP_PORT(default:3000)MCP_HTTP_PATH(default:/mcp)
MCP Client Config (http, default)
{
"mcpServers": {
"xrpl-data": {
"transport": {
"type": "http",
"url": "http://127.0.0.1:3000/mcp"
}
}
}
}
MCP Client Config (stdio, optional)
{
"mcpServers": {
"xrpl-data": {
"command": "node",
"args": ["/Users/ashraychowdhry/Documents/xrpl-data-mcp/src/server.js"],
"env": {
"LOS_BASE_URL": "https://los.prod.ripplex.io",
"DATA_XRPL_BASE_URL": "https://data.xrpl.org",
"XRPL_RPC_URL": "https://s1.ripple.com:51234",
"XRPLMETA_BASE_URL": "https://s1.xrplmeta.org"
}
}
}
}
Response Formats
1) Low-level passthrough tools
Most low-level tools (los_*, vh_*, xrpl_*, validator_history_get, xrpl_public_api_call) return the upstream JSON payload in MCP text content.
Typical shapes:
- XRPL RPC tools:
{ "result": { ... }, "status": "success", "warnings"?: [...] } - VHS/LOS GET tools: service-native object/array response
2) Agent-first composite tools
These return a normalized envelope:
{
"data": {},
"sources": [{ "system": "LOS|VHS|rippled|XRPLMeta", "method": "...", "at": "..." }],
"freshness": { "asOfLedger": 0, "asOfTime": "..." },
"warnings": []
}
3) Error format
On upstream/API/runtime failure, tools return MCP error content with isError: true and human-readable text (often including upstream HTTP status/body).
How To Use Tools
Call MCP tools with:
name: tool idarguments: JSON object matching the tool input schema below
Example:
{
"name": "xrpl_account_info",
"arguments": {
"account": "rHb9CJAWyB4rj91VRWn96DkukG4bwdtyTh",
"ledger_index": "validated"
}
}
How To Use This Server (Step-by-step)
- Install dependencies.
npm install - Start the server locally.
npm start - Add the MCP server config to your MCP client (use the
httpconfig above by default). - In your MCP client, run
tools/listto discover all available tools. - Start with a simple connectivity check:
xrpl_server_infonetwork_overview
- For entity-specific lookups, use
resolve_entitiesfirst, then call suggested tools. - For topology-specific calls:
- call
vh_topology_nodesfirst - take a
pubkeyfrom that response - call
vh_topology_nodewith thatpubkey
- call
- For token/NFT-specific calls, discover live IDs first:
- token:
los_get_trusted_tokens - nft:
xrpl_nft_info/xrpl_nft_historyafter discovering annft_id
- token:
- Run the full validation suite to verify end-to-end health:
node scripts/test-all-tools.mjs > /tmp/xrpl-mcp-test-results.json - Manually call a single tool over HTTP:
node scripts/test-http-tool.mjs http://127.0.0.1:3000/mcp xrpl_server_info '{}'
Full Tool Catalog (49)
Agent-first composite tools (12)
| Tool | Description | Required arguments | Expected response format |
|---|---|---|---|
network_overview |
Network identity, health summary, LOS freshness | none | Envelope: data includes server/network/health metrics |
ledger_summary |
Canonical ledger facts + LOS hints | none (ledger_index/ledger_hash optional) |
Envelope: data.ledger, ledger metadata, optional LOS artifacts |
tx_explain |
Normalized transaction explanation/classification | tx_hash |
Envelope: tx details, classification, token/entity hints |
account_overview |
Account state/activity summary | account |
Envelope: balances, trustlines, activity histogram, risk indicators |
token_overview |
Consolidated issued-token view | issuer, currency |
Envelope: token metadata, holders/trustlines, liquidity/activity |
market_snapshot |
Orderbook + AMM + recent LOS trades | base, quote |
Envelope: orderbook summary, AMM state, trade sample/VWAP |
amm_overview |
AMM state and swap activity | none (amm_id or assetA+assetB recommended) |
Envelope: AMM state, swap sample, aggregate volume |
validator_set_overview |
Validator set composition summary | none (options.group optional) |
Envelope: validator count, operator concentration, set sample |
validator_health |
Validator reliability metrics | pubkey_or_node |
Envelope: validator profile + signed/missed/uptime-like metrics |
amendment_status |
Amendment enablement + context | none (network optional) |
Envelope: enabled amendments, vote context, network context |
search_transactions |
Filtered LOS transaction search + aggregates | filters |
Envelope: results, cursor, aggregate stats |
resolve_entities |
Pattern-based XRPL entity resolver | input |
Envelope: resolved entity type + suggested next tools |
LOS tools (4)
| Tool | Description | Required arguments | Expected response format |
|---|---|---|---|
los_get_token |
Fetch one LOS token by tokenID (currencyHex.issuer) |
tokenID |
LOS token object |
los_batch_get_tokens |
Batch fetch LOS tokens | tokenIds (array) |
LOS batch token response |
los_get_trusted_tokens |
List trusted/KYC tokens | none | LOS trusted token list (count, tokens) |
los_get_transactions |
Query LOS transactions with paging/sort | none (token strongly recommended) |
LOS transaction query response (results/cursor fields) |
Validator history tools (15)
| Tool | Description | Required arguments | Expected response format |
|---|---|---|---|
vh_list_networks |
List tracked networks | none | VHS networks payload (result, networks) |
vh_topology_nodes |
Topology nodes for all/specific network | none (network optional) |
VHS topology nodes payload |
vh_topology_node |
Topology for a specific node pubkey | pubkey |
VHS topology node payload |
vh_list_validators |
List validators (optionally by group) | none (group optional) |
VHS validators payload |
vh_get_validator |
Validator detail | pubkey |
VHS validator profile |
vh_get_validator_manifests |
Validator manifest history | pubkey |
VHS manifest history payload |
vh_get_validator_reports |
Validator reports | pubkey |
VHS validator reports payload |
vh_get_daily_validator_reports |
Daily validator reports collection | none | VHS daily report payload |
vh_get_amendments_info |
Amendment metadata | none | VHS amendments info payload |
vh_get_amendment_info |
Amendment info by name/id | amendment |
VHS amendment detail payload |
vh_get_amendments_vote |
Amendment votes for a network | network |
VHS amendment votes payload |
vh_get_amendment_vote |
Single amendment vote status | network, identifier |
VHS amendment vote detail payload |
vh_health |
VHS health summary | none | VHS health payload |
vh_metrics |
VHS Prometheus metrics | none | Metrics text/structured payload returned as text |
validator_history_get |
Generic VHS GET passthrough | path |
Raw VHS response for requested path |
XRPL JSON-RPC tools (17)
| Tool | Description | Required arguments | Expected response format |
|---|---|---|---|
xrpl_account_info |
Account root info | account |
XRPL RPC result object |
xrpl_account_objects |
Objects owned by account | account |
XRPL RPC result object |
xrpl_account_lines |
Trust lines | account |
XRPL RPC result object |
xrpl_account_tx |
Account tx history | account |
XRPL RPC result object |
xrpl_ledger |
Ledger by hash/index | none | XRPL RPC result object |
xrpl_ledger_data |
Raw ledger state pages | none | XRPL RPC result object |
xrpl_ledger_entry |
Single ledger entry | none (one locator required by XRPL) | XRPL RPC result object |
xrpl_tx |
Transaction by hash | transaction |
XRPL RPC result object |
xrpl_book_offers |
Orderbook offers | taker_gets, taker_pays |
XRPL RPC result object |
xrpl_amm_info |
AMM pool info | asset, asset2 |
XRPL RPC result object |
xrpl_nft_info |
NFToken state/metadata | nft_id |
XRPL RPC result object |
xrpl_nft_history |
NFToken ownership/transfer history | nft_id |
XRPL RPC result object |
xrpl_nfts_by_issuer |
NFTs by issuer | issuer |
XRPL RPC result object |
xrpl_server_info |
Server status/validated range | none | XRPL RPC result object |
xrpl_fee |
Fee metrics | none | XRPL RPC result object |
xrpl_public_api_call |
Generic JSON-RPC method call | method |
XRPL RPC result object |
xrpl_list_recommended_methods |
Curated method list exposed by dedicated tools | none | { "methods": [...] } |
XRPLMeta tools (1)
| Tool | Description | Required arguments | Expected response format |
|---|---|---|---|
xrplmeta_get |
Generic XRPLMeta GET passthrough | path |
Envelope: data contains XRPLMeta response |
Input Discovery Notes (for reliable testing)
Some tools need live identifiers. The test harness discovers them first:
vh_topology_node.pubkey: taken fromvh_topology_nodesoutputvh_get_validator*/validator_health: taken fromvh_list_validatorstx_explain/xrpl_tx: tx hash fromxrpl_account_txxrpl_nft_info/xrpl_nft_history: NFT id from recent NFT transactionslos_get_token/ token-dependent tools: token fromlos_get_trusted_tokens
Test Results
Latest end-to-end live run (2026-02-20):
- Tools tested:
49 - Passed:
49 - Failed:
0
Run command:
node scripts/test-all-tools.mjs > /tmp/xrpl-mcp-test-results.json
Detailed output:
/tmp/xrpl-mcp-test-results.json/Users/ashraychowdhry/Documents/xrpl-data-mcp/docs/TEST_RESULTS.md
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。