mcpField
P
Prax Data Services

Prax Data Services

Remote
mcp.orabrain.comSecurity0Updated 1d ago

About Prax Data Services

Prax Data Services is a financial security and compliance MCP server for AI agents. It screens wallets, entities, and transfers, analyzes transactions, and provides threat-intelligence alerts with reasons and references for its verdicts.

Key features

  1. 01Wallet and smart-contract risk screening
  2. 02Sanctions checks for people and entities
  3. 03Pre-signature transfer screening
  4. 04Transaction risk analysis
  5. 05Threat-intelligence alerts

Use cases

  1. 01Check a wallet before sending funds
  2. 02Screen a company against sanctions
  3. 03Review a transfer before signing
  4. 04Monitor new exploit alerts

MCP support

Exposes screening, transaction analysis, search, and alert tools through a Streamable HTTP MCP endpoint. It supports OAuth 2.1 sign-in or bearer-token authentication.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.orabrain.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

Prax Data Services — MCP server

Prax Data Services MCP Registry

Security and compliance intelligence for financial services, as tools your AI agents can call: wallet / smart-contract risk screening (KYT / AML), pre-signature transfer screening, sanctions screening of people, companies, vessels and crypto addresses, transaction analysis and a threat-intelligence alert feed. Verdicts come with reasons and references — no black-box score.

Endpoint: https://mcp.orabrain.com/mcp (Streamable HTTP, OAuth 2.1 — no key to paste) Docs & plans: https://mcp.orabrain.com · Free plan: 10 credits/day, no card.

Connect

ClientHow
Claude (claude.ai / desktop)Settings → Connectors → Add custom connector → URL https://mcp.orabrain.com/mcp → Connect (sign in on the portal, approve)
Claude Codeclaude mcp add --transport http prax https://mcp.orabrain.com/mcp then /mcp → prax → Authenticate
CursorAdd to Cursor or add to .cursor/mcp.json: {"mcpServers":{"prax":{"url":"https://mcp.orabrain.com/mcp"}}}
VS Code (Copilot agent mode).vscode/mcp.json: {"servers":{"prax":{"type":"http","url":"https://mcp.orabrain.com/mcp"}}}
ChatGPTSettings → Connectors → Create (Developer mode) → MCP server URL above, authentication OAuth
Anything elseAny MCP client with Streamable HTTP + OAuth (dynamic client registration supported), or send Authorization: Bearer prax_live_… with a key from the portal

The first call sends you to mcp.orabrain.com/portal to sign in (or create a free account) and approve the app. Connected apps are listed under Account → Connected apps, each with its own key; disconnect any time.

Tools

ToolWhat it answersPlan
screen_walletIs this EVM address safe? Sanctions, exploits, phishing, scams, drainers, mixers, malicious contracts, first-hop exposure. Risk band, labels, roles, alerts, counterparties, plain-language story.Free
screen_entitySanctions screening of a person, company, vessel or address (fuzzy names, aliases, linked addresses).Free
screen_transferPre-signature check: allow / review / block for amount of asset from A to B, with reasons. No chain call.Builder
analyze_transactionStatus, value, counterparties and risk of both sides for a tx hash.Builder
screen_wallets_batch / screen_transfers_batchUp to 50 per call — portfolio sweeps, payout files.Builder
search / fetchExplore Prax intel in plain words; fetch a full record.Business
get_alertsApproved threat-intel alerts from the last 24 h (same payload as the webhook feed).Business

Try: "Is 0x5227f603C1eE34CE3C06eFe6477003BDfCCB2f8c safe to pay?" · "Screen Acme Trading Ltd against sanctions" · "Any new exploit alerts today?"

REST

Same backend, same metering: https://mcp.orabrain.com/v1 — OpenAPI at /v1/openapi.json. See examples/.

Terms

Output is licensed for use inside your own product only — no resale, redistribution, model training or competing services. Terms of Service · Privacy

Registry

Published to the official MCP Registry as com.orabrain/prax-data-services (see server.json).

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.