xrpldashboard

Official15 toolsLive

by Enkryptedx · Python

Read-only XRP Ledger MCP tools with proof-annotation envelopes and signed daily snapshots.

Read-only Model Context Protocol (MCP) tools for the XRP Ledger, designed for public analytics. The server exposes tools that deliver proof-annotation envelopes and includes signed daily snapshots. The repository emphasizes auditability for headline numbers, with claim records and automated checks before changes are pushed.

🛠️ Key Features

Read-only MCP tools for XRP Ledger analytics
Proof-annotation envelopes for tool outputs
Signed daily snapshots
Truth-audit system using CLAIMS.yaml
scripts/claims_check.sh runs before pushes

🚀 Use Cases

Access public analytics for XRP Ledger endpoints (e.g., /rlusd, /whales, /coverage, /analytics)
Validate freshness and source linkage of reported numeric claims
Use daily signed snapshots for recurring reads

⚡ Developer Benefits

Claim records enumerated in CLAIMS.yaml and tied to source functions
Four-layer audit design documented in docs/TRUTH_AUDIT_DESIGN.md
Pre-push verification via scripts/claims_check.sh

⚠️ Limitations

Intended for public analytics only (read-only tools)

Topics

blockchain-analyticsdashboarddata-transparencyflaskpostgresqlpublic-goodspythonxrp-ledgerxrplsigned-snapshots

Captured live from the server via tools/list.

get_ledger_stats

Return the last-validated ledger stats (index, close time, server state, load factor, complete range, build). Wrapped in the xrpldashboard proof envelope; envelope source is derived from the actual URL host (`local_rippled` for loopback / private LAN, `remote_rippled:<host>` otherwise) — never hardcoded.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_ledger_statsArguments"
}

get_amendment_status

Return the current amendments state (enabled / in-flight / superseded / unrecognized-enabled) with `cross_check_status` derived from the responding node's feature-RPC vs Amendments ledger-object concurrence. Wrapped in the proof envelope.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_amendment_statusArguments"
}

get_unl_status

Return the current UNL composition (both published lists, overlap, expiry-days). Honest-partial when a list fetch fails. Wrapped in the proof envelope.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_unl_statusArguments"
}

get_whale_events

Return the last N XRP-denominated whale events + tagged-watchlist events (trustset excluded). Same feed the homepage globe pulse reads. Wrapped in the proof envelope.

Parameters1

limit

integer

optional

Raw schema

{
  "type": "object",
  "properties": {
    "limit": {
      "default": 25,
      "title": "Limit",
      "type": "integer"
    }
  },
  "title": "get_whale_eventsArguments"
}

get_whale_watchlist

Return the last N tagged-watchlist events (type='tagged'), floored at TAGGED_XRP_FLOOR_DROPS for XRP-denominated rows. Wrapped in the proof envelope.

Parameters1

limit

integer

optional

Raw schema

{
  "type": "object",
  "properties": {
    "limit": {
      "default": 25,
      "title": "Limit",
      "type": "integer"
    }
  },
  "title": "get_whale_watchlistArguments"
}

get_rlusd_supply

Return the current cross-chain RLUSD supply (Ethereum + XRPL) from the same PG mirror /rlusd reads. Honest-partial when either chain errored. Wrapped in the proof envelope.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_rlusd_supplyArguments"
}

get_rlusd_flow_24h

Return the last finalized 24h XRPL net-change from rlusd_supply_history. freshness_contract='finalized_only' is the machine-readable form of the R1/R2 finalized-window rule. cross_check_status derived from stored vs supply-diff pair. Wrapped in the proof envelope.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_rlusd_flow_24hArguments"
}

get_amm_pool

Return the single ranked-pool row for one AMM account from the amm_ranked_pools snapshot. Raises when the account is absent (unranked, below thresholds, or newer than the last snapshot). Wrapped in the proof envelope.

Parameters1

amm_account

string

required

Raw schema

{
  "type": "object",
  "properties": {
    "amm_account": {
      "title": "Amm Account",
      "type": "string"
    }
  },
  "required": [
    "amm_account"
  ],
  "title": "get_amm_poolArguments"
}

get_amm_top_by_tvl

Return the top-N ranked AMM pools by tvl_usd from the same snapshot /pools reads. NULL-tvl rows sort last. Wrapped in the proof envelope.

Parameters1

limit

integer

optional

Raw schema

{
  "type": "object",
  "properties": {
    "limit": {
      "default": 10,
      "title": "Limit",
      "type": "integer"
    }
  },
  "title": "get_amm_top_by_tvlArguments"
}

get_token_attestation

Return the attestation tier (verified / self-described / null) for one (currency, issuer) pair. Data payload includes `dispute_contact_url` — first tool to name a third party. Wrapped in the proof envelope.

Parameters2

currency	string	required
issuer	string	required

Raw schema

{
  "type": "object",
  "properties": {
    "currency": {
      "title": "Currency",
      "type": "string"
    },
    "issuer": {
      "title": "Issuer",
      "type": "string"
    }
  },
  "required": [
    "currency",
    "issuer"
  ],
  "title": "get_token_attestationArguments"
}

get_rwa_families

Return every rwa_family row with attributed pool count. Data payload includes `dispute_contact_url`. Wrapped in the proof envelope.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_rwa_familiesArguments"
}

get_rwa_pools

Return every attributed RWA pool with provenance + TVL. Data payload includes `dispute_contact_url`. Wrapped in the proof envelope.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_rwa_poolsArguments"
}

get_mpt_snapshot

Return the MPT snapshot dict (issuances / by_class / total). Not third-party-naming — no dispute_contact_url. Wrapped in the proof envelope.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_mpt_snapshotArguments"
}

get_signed_snapshot

Return the Ed25519-signed daily snapshot for `date_str` (ISO YYYY-MM-DD) wrapped in the proof envelope. Absence raises — the walker has not produced this date yet, or the date predates the chain start.

Parameters1

date_str

string

required

Raw schema

{
  "type": "object",
  "properties": {
    "date_str": {
      "title": "Date Str",
      "type": "string"
    }
  },
  "required": [
    "date_str"
  ],
  "title": "get_signed_snapshotArguments"
}

verify_snapshot_signature

Stateless verification: re-derive the leaf hash, check the Ed25519 signature against the pinned pubkey, check the audit path against the claimed chain root. Accepts full MCP envelope or bare signed payload. Wrapped in the proof envelope.

Parameters1

envelope

object

required

Raw schema

{
  "type": "object",
  "properties": {
    "envelope": {
      "additionalProperties": true,
      "title": "Envelope",
      "type": "object"
    }
  },
  "required": [
    "envelope"
  ],
  "title": "verify_snapshot_signatureArguments"
}

README.md7 KBView on GitHub

xrpldashboard

Public XRPL analytics with a built-in truth-audit system. Live at xrpldashboard.com.

What this is

Three things make this repo different from "another blockchain dashboard":

Every headline number has a claim record. CLAIMS.yaml enumerates the public numeric claims across /rlusd, /whales, /coverage, and /analytics, each tied to its source function and freshness contract. scripts/claims_check.sh runs before pushes.
A four-layer audit catches stale numbers before readers do. See docs/TRUTH_AUDIT_DESIGN.md. Retro-tested against the last nine real incidents: 9 of 9 caught. The first live catch (an RLUSD partial-day bug) surfaced within hours instead of the 53 days it had previously gone unnoticed.
Daily signed snapshots. Ed25519-signed, Merkle-chained. /signed-snapshots and .well-known/snapshots/<date>.json let anyone verify a number as-of a date without trusting the site.

Quick links: /methodology · /signed-snapshots · docs/TRUTH_AUDIT_DESIGN.md

Agent tier (MCP)

Live public MCP endpoint: https://mcp.xrpldashboard.com/mcp — streamable-HTTP, protocol version 2025-06-18, no auth. Fifteen read-only tools over XRPL and the on-XRPL RLUSD supply, every response wrapped in a proof-annotation envelope with source, as_of, freshness_contract, and a claims_ref back to /claims. Public beta through 2026-09.

Connect in 60 seconds — copy-paste config for Claude Desktop or any MCP client:

json

{
  "mcpServers": {
    "xrpldashboard": {
      "command": "npx",
      "args": [
        "mcp-remote@latest",
        "https://mcp.xrpldashboard.com/mcp?ref=readme"
      ]
    }
  }
}

Or add it as a Custom Connector in Claude Desktop → Settings → Connectors with URL https://mcp.xrpldashboard.com/mcp and auth None. Full onboarding page (three sample prompts, honest limits, 429 shape): /connect.

Session rate limit: 600 tool calls / hour / session, enforced live (HTTP 429 + Retry-After on breach). See mcp_session_rate_limit.py.
Discovery manifest: /.well-known/agents.json · /llms.txt · /openapi.json.
Verifiable moat: get_signed_snapshot + verify_snapshot_signature — pin the Ed25519 pubkey at /.well-known/snapshots/pubkey.pem and verify a day's snapshot without trusting us.
Backing infra: our own rippled full-history node (Ubuntu box in Indiana → Cloudflare Tunnel), source at mcp_server.py and mcp_tools_*.py.

Design doc: docs/AGENT_TIER_DESIGN.md.

Architecture

Web: Flask 3.1 + Jinja2 + Flask-Babel (i18n) + Flask-Limiter, deployed on Render, fronted by Cloudflare.
Data: Neon Postgres (single shared DB across web + walkers).
Ingest: 29 background services under launchd — 23 ingest walkers and canaries, plus 4 backup and 2 snapshot-signing jobs. Each walker writes to walker_health; /walker-health surfaces stalls.
XRPL client: xrpl-py 4.5.0 with a local-first cascade to public rippled nodes; silent-failover attempts logged to walker_node_fallback so we can audit reliability rather than hope.
Signing: Ed25519 via cryptography (see signed_snapshot.py).

The truth-audit system (why this repo may be reusable)

Most public dashboards trust their own writes. This one doesn't. Four layers, working together:

Layer 1 — walker health. walker_health rows + /walker-health page. Catches "the number stopped moving because the writer died."
Layer 2 — plausibility rules. Continuous checks like "24h net-change should not equal zero for 53 days." Catches "the number is moving in the DB but the query is wrong."
Layer 3 — external cross-check. Third-party comparisons (e.g., independent Ethereum RPCs for RLUSD supply, CoinGecko for XRP price) surface disagreement on the same measurement.
Layer 4 — claims manifest. CLAIMS.yaml + scripts/claims_check.sh. Every headline number is enumerated with its source function and freshness contract; the checker exits non-zero on drift.

Full design and the 9-of-9 retro-test are in docs/TRUTH_AUDIT_DESIGN.md.

If you're building a public-data project and want the pattern, the design doc is written to be lifted.

Running locally

What runs in five minutes: the Flask web app against a copy of the schema. Enough to click through pages and see the UI.

What does not run in five minutes: the full site, because a live dashboard needs the walker fleet ingesting from XRPL against Neon Postgres. Standing up the walkers is a real operations task, not a docker compose up.

Quickstart (web app only):

bash

python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env      # then fill in DATABASE_URL, FLASK_SECRET_KEY, etc.
python app.py             # http://localhost:5001

For the full ingest setup, see docs/TRUTH_AUDIT_DESIGN.md and the launchd/ plists (which are the author's local install — forkers will want to path-adjust).

Independence

xrpldashboard is not operated by Ripple, the XRPL Foundation, or any exchange. No paid placements. No affiliate links on labeling or metric surfaces.

Funding

Development is self-funded, with support from community grant programs where they align with the mission (public-goods data infrastructure for XRPL).

Contributing

Issues and PRs welcome. If you're touching a page that renders a headline number, add or update the corresponding entry in CLAIMS.yaml; scripts/claims_check.sh will remind you.

License

MIT. See LICENSE.

Disclaimer

xrpldashboard reports on-chain data and public regulatory information. Nothing on the site or in this repo is financial, legal, or investment advice.

Install

Tools (15)

get_ledger_statsget_amendment_statusget_unl_statusget_whale_eventsget_whale_watchlistget_rlusd_supplyget_rlusd_flow_24hget_amm_pool

Remote endpoint

Streamable HTTP

Hosted server - connect over the network, no local install.

url

https://mcp.xrpldashboard.com/mcp?ref=anthropic

Add to VS Code Add to Cursor

claude_desktop_config.json

json

{
  "mcpServers": {
    "xrpldashboard-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.xrpldashboard.com/mcp?ref=anthropic"
      ]
    }
  }
}

Desktop config is stdio-only; this bridges via mcp-remote. Native remote: Settings > Connectors.

xrpldashboard

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

Topics

get_ledger_stats

get_amendment_status

get_unl_status

get_whale_events

get_whale_watchlist

get_rlusd_supply

get_rlusd_flow_24h

get_amm_pool

get_amm_top_by_tvl

get_token_attestation

get_rwa_families

get_rwa_pools

get_mpt_snapshot

get_signed_snapshot

verify_snapshot_signature

xrpldashboard

What this is

Agent tier (MCP)

Pages

Architecture

The truth-audit system (why this repo may be reusable)

Running locally

Independence

Funding

Contributing

License

Disclaimer

Install

Remote endpoint

Related servers