Give your AI agent verification as a tool
Verifisha’s MCP server lets an AI assistant or agent run Kenyan identity, business, CRB, AML and good-conduct checks mid-task — consent-led, built on the same API your app uses. It’s in preview; request access to connect your agent.
Request preview accessTell us about your agent and we’ll onboard you. No spam.
Verification, exposed as MCP tools
verify_identityConfirm a Kenyan national ID against IPRS — name match, ID status and biographic fields.
verify_businessLook up a company on the BRS for KYB — registered name, number and director signals.
crb_checkPull Credit Reference Bureau status as part of a lending, tenant or hiring decision.
aml_screenScreen a person or director for sanctions, PEP and adverse-media risk.
request_referencesSend structured reference requests and collect referee responses.
get_reportFetch a completed check or a reusable verified profile with its consent record.
Tool names are illustrative of the preview and may change before general availability.
Agent-friendly, without cutting corners
Consent-led by design
Every check an agent runs is still consent-led under Kenya’s Data Protection Act. The subject approves, and the approval is stored as evidence — agents do not bypass consent.
Built on the Verifisha API
The MCP server is a thin, typed wrapper over the same REST API your developers already use, so agent calls and app calls behave identically.
Reusable verified profiles
Agents can reuse an existing verified profile instead of re-running checks — faster, cheaper and less intrusive.
Pay per check
The same pay-per-check pricing applies — no subscription — so agent-driven verification scales with use.
Frequently asked questions
What is the Verifisha MCP server?
It is a Model Context Protocol (MCP) server that exposes Verifisha’s checks — identity (IPRS), business (BRS), CRB, AML/PEP, good-conduct and references — as tools an AI agent can call. An assistant like Claude can verify a person or business mid-conversation, with consent, and get a structured result back.
What is MCP (Model Context Protocol)?
MCP is an open standard for connecting AI assistants and agents to external tools and data. An MCP server advertises a set of tools; an AI client can then call them safely. Verifisha’s MCP server makes verification one of those tools.
Is the MCP server available now?
It is in preview. The underlying REST API is live today, and the MCP wrapper is being rolled out with early partners. Request access and we’ll onboard you and share the connection details.
Does an AI agent skip consent?
No. Agent-run checks are consent-led exactly like checks run in the app — the subject approves and the approval is recorded. MCP changes how the check is triggered, not the lawful basis for it.
How is it authenticated and billed?
Through your Verifisha account and API credentials, with the same pay-per-check pricing. You keep full visibility and control over what agents can run.
Verify anyone — or get verified yourself.
Run consent-led KYC, KYB, AML, CRB and background checks in one place — or build your own verified, shareable trust profile for jobs, rentals and deals. Trust made simple, for Kenya and Africa.
Create your account and start in minutes — no sales call or demo needed. Talking to sales is optional.