Use case
Compliance tools for AI agents
AI agents that touch payments, onboarding or counterparties need the same screening checkpoints as application code. CompliAPI exposes its full screening surface as MCP tools agents can call natively.
As operational work shifts to AI agents — onboarding assistants, payment-ops automations, support copilots — the compliance checkpoints that lived in application code need to be available where agents work: as tools.
CompliAPI ships a hosted MCP server that exposes every screening capability as a tool over streamable HTTP: wallet, email, website and government-ID screening against OFAC SDN data, country checks, geolocation, and VPN detection. Agents authenticate with your existing API token, and every tool call is metered and logged like a REST request.
Where CompliAPI fits
Screening as a tool call
Agents call tools like screen_crypto_address and screen_email directly, and get the same structured results your application code would.
OFAC MCP server for AI agents →Counterparty checks before actions
An agent preparing a payout screens the destination wallet and the counterparty's website before proposing the action.
Wallet screening API for crypto compliance →Deterministic backstops
Critical flows keep an enforced REST-API check in application code — same data, same credentials — so screening never depends on the agent alone.
OFAC API — screen against the official SDN list →A reviewable trail
Tool calls land in your organization's request log alongside REST traffic, so you can review exactly what your agents screened.
OFAC MCP server for AI agents →Scope
MCP tools expose the same screening scope as the REST API: direct checks against OFAC SDN data plus geolocation and VPN signals. An agent's tool result is an input to its reasoning, not an enforcement mechanism — keep deterministic checks in critical paths. CompliAPI supplies the tools, not legal advice.
Frequently asked questions
Which agents and frameworks can use the MCP server?
Any MCP client that supports remote servers over streamable HTTP — add the server URL and an Authorization bearer header to the client's configuration.
Do agent tool calls need separate credentials?
No. The MCP server authenticates with the same API tokens as the REST API, and tool calls count against the same organization quota.
How do I audit what an agent screened?
Every tool call is recorded in your organization's request log exactly like a REST request, so the trail covers human and agent traffic alike.
Should screening decisions be left to the agent?
For critical flows, no — most teams enforce checks deterministically in application code and give agents the same tools for context and triage. A tool result informs the agent; your application enforces the policy.
The solutions behind this use case
AI infrastructure
OFAC MCP server for AI agents
Give AI agents sanctions screening tools over the Model Context Protocol.
Sanctions screening
OFAC API — screen against the official SDN list
Screen crypto wallets, emails, IDs and countries against the official US Treasury OFAC SDN list, refreshed every 15 minutes.
Give your agents screening tools
Get a free API key and connect the MCP server in minutes.
14-day free trial. No credit card required.