Kredit
Kredit provides an identity verification and risk intelligence layer for AI agents conducting payments and API calls. The platform issues verifiable agent passports tied to human and business identities, and its Check service analyzes nine risk layers to return allow, escalate, or deny decisions in as little as 6 milliseconds. Kredit integrates via REST API, MCP, CLI, and SDKs, with human-in-the-loop approval for sensitive actions.
- Regulatory & Compliance Technology
- Financial Technology
- Cybersecurity
- AI Agents
- Artificial Intelligence
- Software Only
Funding
Founders
Founder details are not available yet.
Product
Problem
As AI agents increasingly execute purchases, API calls, and payments autonomously, businesses lack a reliable way to verify who is acting on their behalf, confirm transaction mandates, and assess risk in real time. Unverified agents can commit fraud, exceed spending limits, or interact with untrusted counterparties, creating significant financial and compliance exposure.
Solution
Kredit provides an identity and risk intelligence layer for agentic commerce, issuing verifiable "passports" that tie every AI agent to a human (KYC), business (KYB), and mandate, with an expiry and full audit record. Before each transaction, its "check" service analyzes nine layers of risk—identity, mandate, payee, behavior, and more—across APIs, MCP, CLI, or SDK, returning an allow, escalate, or deny decision in as little as 6 milliseconds. Only verified humans can register agents or change their mandates, and sensitive actions (new rules, high-value payments) require explicit human approval in the console or iOS app. The system surfaces suspicious activity in a dashboard for review while enabling automated enforcement of per-agent and per-org rules, spending limits, and rate limits. A unified API powers both the console and all integration surfaces, with human-in-the-loop approval for changes that exceed pre-set thresholds.
Target Audience
Kredit targets security, finance, and procurement teams at companies deploying AI agents for purchasing, payment processing, or API-driven operations, as well as merchants and payment service providers that need to authenticate agent-initiated transactions. It serves both buyers implementing agentic commerce and the payment/service platforms that must verify these actors.
Features
- Kredit Passport: KYA, KYB, and KYC-verified agent identity with mandates, expiration, activity records, and dispute counts
- Kredit Check: nine-layer risk scoring (identity, mandate, payee, behavior) returning allow/escalate/deny in milliseconds
- Human-in-the-loop controls: fingerprint-verified human registration and approval for agent changes, new rules, and payments above review thresholds
- Multi-surface integrations: REST API (POST /check), MCP server, CLI installer, and SDKs for Python and Node.js
- Comprehensive rules engine: payment/api/tool rules with spend limits, rate limits, allowed/blocked lists, and deny-or-review hit policies
- Guardrails for agent execution: per-action ceilings, merchant list controls, and per-agent or per-organization spend windows
- Console dashboard with loss prevention tracking, agent management, transaction history, escalation queues, and wallet monitoring
- x403 protocol: a signed risk decision protocol for agentic payments, with portable receipts verifiable offline