Capri Ai Agentpay

Capri AgentPay is the first governed payment layer purpose-built for AI agents. It solves a critical problem that teams face when their agents start interacting with paid APIs and settling transactions autonomously. Handing an agent a private key is a security nightmare because a single bad prompt or unexpected behavior can drain an entire wallet. Capri AgentPay puts a secure runtime between the agent and the signer, so the agent never directly controls your keys or funds.

When an agent makes a request and hits an HTTP 402 Payment Required challenge, Capri AgentPay intercepts that request and evaluates it against a fixed policy order. The policy checks include a merchant allowlist, a per-transaction cap, a daily budget limit, and an auto-approve threshold that can pause for human review. Only when every single check passes does the system settle the payment, retry the original request with proof attached, and store a signed receipt that you can verify offline. This means your agents can call paid APIs and MCP tools safely, while you maintain full control over spending and approvals.

Capri AgentPay is designed for engineering teams shipping agent systems in production. It provides six MCP tools instead of raw key access, a VS Code extension for easy integration, and a Base Sepolia sandbox for testing. The product gives operators clear review queues, audit trails, and controls that do not collapse back into manual busywork. Whether you are building research agents, automation pipelines, or customer-facing AI tools, Capri AgentPay gives you the confidence to let your agents touch real money and live infrastructure.

HTTP 402 Payment Handling

Capri AgentPay natively handles HTTP 402 Payment Required challenges that agents encounter when calling paid APIs. When an agent hits a 402 response, the system intercepts it, evaluates your configured policy, and only settles the payment if all checks pass. It then retries the original request with proof of payment attached, so the agent gets the data it needs without any manual intervention.

Per-Agent Budgets and Caps

You can set granular budgets and spending caps for each individual agent. This includes per-transaction limits and daily budget constraints that prevent any single agent from overspending. The system enforces these limits in a fixed order, giving you predictable and auditable control over every payment your agents make.

Merchant Allowlists

Capri AgentPay lets you define allowlists of approved merchants or API endpoints that your agents can pay. Any request to a merchant not on the allowlist is automatically denied, providing a strong first line of defense against unauthorized spending. This ensures your agents only interact with trusted services.

Human-in-the-Loop Approvals

When a payment request exceeds your auto-approve threshold, Capri AgentPay pauses the transaction and requests human review. Operators see the exact amount, asset, chain, and recipient before approving or denying the payment. This gives you confidence that large or unusual transactions are always reviewed by a person.

Verifiable Receipts

Every settled payment generates a signed receipt that you can verify offline. The receipt includes hashes, timestamps, and verification fields, providing a complete audit trail for every transaction. This is invaluable for compliance, accounting, and security reviews.

Developer-First Tooling

Capri AgentPay provides six MCP tools instead of raw key access, a VS Code extension for seamless integration, and a Base Sepolia sandbox for testing. The developer experience is designed to be small and easy to reason about, with typed challenge requirements, explicit policy order, and clear documentation.

Autonomous Research Agents

Research agents that need to call premium data APIs or paid knowledge bases can use Capri AgentPay to handle payments autonomously. The agent requests data, Capri evaluates the policy, and if approved, the payment is settled and the data is returned. This allows research teams to deploy agents that access paid resources without manual payment steps.

Automated Infrastructure Management

Teams running automated infrastructure agents that need to call paid cloud APIs or third-party services can use Capri AgentPay to control spending. Each agent has its own budget and caps, and any unusual spending triggers a human review. This prevents runaway costs while still allowing agents to operate independently.

Customer-Facing AI Tools

When building customer-facing AI products that rely on paid APIs for features like image generation, natural language processing, or data enrichment, Capri AgentPay ensures that agent spending stays within budget. The merchant allowlist prevents agents from calling unauthorized services, and the audit trail provides full visibility into costs.

Internal Automation Pipelines

Internal automation agents that handle tasks like data processing, report generation, or system integration often need to call paid APIs. Capri AgentPay gives these agents a secure payment path while keeping your keys safe. Operators can review transactions, set budgets per department or project, and get signed receipts for accounting.

How does Capri AgentPay protect my private keys?

Capri AgentPay puts a runtime between your agent and the signer. The agent never holds your private keys. Instead, it calls a typed surface, and Capri evaluates your policy before any signature is created. This means even if an agent is compromised or given a bad prompt, your keys and wallet remain safe.

What happens if a payment exceeds my budget?

If a payment request exceeds your per-transaction cap or daily budget, Capri AgentPay denies the transaction automatically. If it exceeds your auto-approve threshold but is within your budget, the system pauses and requests human review. Operators can then approve or deny the payment, giving you full control over spending.

Can I test Capri AgentPay before using it in production?

Yes, Capri AgentPay provides a Base Sepolia sandbox for testing. You can set up policies, test payment flows, and see how the system handles different scenarios without risking real funds. The VS Code extension also makes it easy to integrate and test locally.

What kind of receipts does Capri AgentPay provide?

Every settled payment generates a signed receipt that includes hashes, timestamps, and verification fields. You can verify these receipts offline, which is useful for compliance, accounting, and security audits. The receipts provide a complete and tamper-proof record of every transaction.