Hive Agent Kyc

io.github.srotzinPublished by GitHub account srotzinv1.0.0Updated Oct 24, 2026

KYA agent identity verification and trust scoring for autonomous A2A networks

VerifiedStreamable HTTPWeb executableFinanceSecurity & Monitoring

Overview

AI-generated overview

Remote MCP broker that lets an assistant screen blockchain addresses through third-party KYC/AML providers and check OFAC and FATF public lists.

What you can do
Screen a blockchain address through Chainalysis, TRM Labs or Elliptic and get the provider's risk score and flags back verbatim (currently returns a backend-pending error until provider keys are configured). Check an identifier against the public OFAC SDN sanctions list, cached for 24 hours. Check an ISO-3166 country code against FATF call-for-action and increased-monitoring lists. Read back an audit-log entry for a prior query by its query ID, returning only DID, timestamp, provider, result code and an address hash.
When to use it
Useful when an agent workflow needs a pre-settlement sanctions or risk check on a blockchain address or country, or a quick lookup against public OFAC and FATF lists. The free public-list tools work today; the paid screening route is not yet operational.
Requirements
A remote Streamable-HTTP endpoint; no local install or account is declared. The screening tool is quoted at $0.10 USDC on Base L2 via x402 settlement. Running it locally needs Node.js and optionally the CHAINALYSIS_API_KEY, TRM_API_KEY or ELLIPTIC_API_KEY environment variables; without them the screening tool returns a backend-pending error.
Before you install
The screening tool is a paid call settled in USDC on Base L2 to a stated wallet address; the README says no charge applies while the backend is pending, but payment configuration should be verified before use. Screening requests are routed to third-party providers. The server is a broker only and makes no compliance determination. Audit entries are held in memory and are not durable.

Installation

In SourceWeft

  1. Open Hive Agent Kyc in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/io.github.srotzin%2Fhive-mcp-agent-kyc/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "hive-agent-kyc": {
      "type": "http",
      "url": "https://hive-mcp-agent-kyc.onrender.com/mcp"
    }
  }
}

README

[Hive Civilization MCP Gateway · broker / observer layer · CLEAN-MONEY gate]

hive-mcp-agent-kyc

Broker / observer layer for KYC/AML screening · CLEAN-MONEY gate primitive

[Smithery] [Glama] [Live] [Release] [License]

https://hive-mcp-gateway.onrender.com/agent-kyc/mcp


Scope disclaimer

Hive Agent KYC is a broker/observer layer. It routes screening requests to third-party providers and surfaces public sanctions list matches. It is not a regulated money-services business, does not make custody, lending, or transaction-permitting decisions, and does not issue compliance attestations on behalf of regulated entities. Final compliance determinations remain with the requesting agent and its operator.

This disclaimer is also surfaced verbatim in every tool's response payload, in /.well-known/agent.json, and in the JSON-LD description.


What this is

hive-mcp-agent-kyc is a Model Context Protocol (MCP) server that any MCP-compatible client (Claude Desktop, Cursor, Manus, agent runtimes) can call to:

  1. Route a blockchain address screening request to one of three third-party KYC/AML providers — Chainalysis, TRM Labs, or Elliptic — and return the provider's risk score and flags verbatim (no enrichment, no Hive scoring layer).
  2. Check public sanctions list matches against the OFAC SDN list and the FATF high-risk-jurisdiction lists.
  3. Read back an audit-log entry for a prior query (DID + timestamp + provider + result code only — no PII).

What this does not do:

  • It does not hold custody of any KYC documents.
  • It does not make a final allow/deny determination on a transaction.
  • It does not issue an attestation that a person/entity is "KYC'd by Hive".
  • It does not operate as a money-services business.

The purpose is to be the CLEAN-MONEY gate primitive that other Hive surfaces (Vault, Trade, Swap, Treasury) can call before settling. Compliance determinations remain with the calling operator.

  • Protocol: MCP 2024-11-05 over Streamable-HTTP / JSON-RPC 2.0
  • Transport: POST /mcp
  • Discovery: GET /.well-known/mcp.json, GET /.well-known/agent.json
  • Health: GET /health
  • Settlement: real x402 — USDC on Base L2 to wallet 0x15184bf50b3d3f52b60434f8942b7d52f2eb436e. No mock. No simulated.
  • Backend status: provider partnership keys pending — agent_kyc_screen_address returns 503 with backend_pending: true until configured. Public list tools (OFAC, FATF) return real data today.
  • Brand gold: Pantone 1245 C / #C08D23

Tools

ToolCostDescription
agent_kyc_screen_address$0.10 USDC (Base)Route screening request to Chainalysis / TRM Labs / Elliptic. Returns provider response verbatim. 503 backend_pending until keys configured.
agent_kyc_check_ofac_listfreeCheck identifier against OFAC SDN public list. Cached 24h.
agent_kyc_check_fatf_listfreeCheck ISO-3166 country code against FATF Call-for-Action and Increased-Monitoring lists.
agent_kyc_query_statusfreeRead prior audit-log entry by query_id. Returns DID + timestamp + provider + result code + address hash. No PII.

agent_kyc_screen_address

json
{  "name": "agent_kyc_screen_address",  "arguments": {    "address": "0x...",    "chain": "base",    "provider": "chainalysis",    "requester_did": "did:hive:agent-vault"  }}

Until partnership keys are configured, returns:

json
{  "ok": false,  "status": 503,  "backend_pending": true,  "message": "Third-party screening backend not yet configured...",  "x402": {    "recipient": "0x15184bf50b3d3f52b60434f8942b7d52f2eb436e",    "chain": "base",    "asset": "USDC",    "amount_usdc": 0.10  }}

agent_kyc_check_ofac_list

json
{  "name": "agent_kyc_check_ofac_list",  "arguments": { "identifier": "0x...", "identifier_type": "address" }}

Returns the cache age, source URL (treasury.gov), and a deterministic match flag against the cached SDN list.

agent_kyc_check_fatf_list

json
{  "name": "agent_kyc_check_fatf_list",  "arguments": { "country_code": "IR" }}

Returns category (call_for_action | increased_monitoring | not_listed) + FATF source URL.

agent_kyc_query_status

json
{  "name": "agent_kyc_query_status",  "arguments": { "query_id": "..." }}

Audit log

The audit log stores only these fields per query, in memory:

  • query_id (UUID)
  • did (requesting agent's DID)
  • ts (ISO-8601 timestamp)
  • provider (chainalysis | trm | elliptic | ofac_public | fatf_public)
  • result_code (HTTP status returned)
  • address_hash (first 16 hex chars of SHA-256 of lowercase identifier)

Raw addresses, names, and PII are never stored. Operators wanting durable logs route them to their own SIEM via webhook (not a Hive responsibility).


Connect

Claude Desktop

json
{  "mcpServers": {    "hive-agent-kyc": {      "url": "https://hive-mcp-gateway.onrender.com/agent-kyc/mcp"    }  }}

Cursor / Manus / generic MCP client

Point your client at:

https://hive-mcp-gateway.onrender.com/agent-kyc/mcp

Streamable-HTTP transport, JSON-RPC 2.0, MCP 2024-11-05.

Local

bash
git clone https://github.com/srotzin/hive-mcp-agent-kyccd hive-mcp-agent-kycnpm installnode server.js# server runs on :3000

To enable the screening provider routing locally, set any of:

bash
export CHAINALYSIS_API_KEY=...export TRM_API_KEY=...export ELLIPTIC_API_KEY=...

When no key is set, agent_kyc_screen_address returns 503 backend_pending. The OFAC and FATF tools work without keys.


Settlement

Real x402 rails. The screening tool quotes:

recipient : 0x15184bf50b3d3f52b60434f8942b7d52f2eb436echain     : Base L2asset     : USDCamount    : $0.10 USDC per screen call

No mock. No simulated. No testnet pretending to be mainnet. While the screening backend is in backend_pending, no charge is applied — the x402 quote is informational so calling agents can pre-flight payment configuration.


Hive doctrine context

This server is the CLEAN-MONEY gate primitive in the Hive Civilization three-gate doctrine (NEED + YIELD + CLEAN-MONEY). Any other Hive surface that touches settlement is expected to call agent_kyc_screen_address and the public-list tools before clearing.

It is intentionally a thin broker: enrichment, scoring, and final decision remain with the third-party provider and the operator, respectively. Hive does not become a regulated MSB by routing.


License

MIT — see LICENSE.


#C08D23 · Hive Civilization · 2026

Hive Civilization Directory

Part of the Hive Civilization — agent-native financial infrastructure.

Brand: #C08D23

Source: README.md at commit e1ce3d1

Tools

0
SourceWeft has not read this server's tools yet.

Version history

1
  1. v1.0.0LatestSep 16, 2026

Score

85% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts