Agentic payments

Govern the payment paths you route through Axiru.

Policy checks for payment requests routed through Axiru. x402 pre-authorization is live; Stripe DAA is shadow-only; USDC on Solana is live, enabled per organization. Review the support matrix below.

Rails we govern

Support and enforcement by rail.

x402 pre-authorization is live. Stripe DAA runs in shadow mode. USDC on Solana is live, enabled per organization. Pre-execution control requires routing the payment through the supported integration.

Stripe DAALive in shadow mode

Direct-Account-Authorized agent payments. Axiru evaluates agent_payment, agent_payout, and agent_authorization in shadow mode: decisions recorded for review. Shadow evaluation does not gate or execute a DAA payment.

x402Pre-authorization live

HTTP 402 payment-required flows for agent-to-agent and human-to-agent commerce. Policy evaluates the payment before it executes and, on allow, issues a signed authorization token verifiable with the published JWKS. An integrated caller must reject a missing or invalid token before settlement.

Stripe (legacy)GA (refunds enforcing)

Refund enforcement pre-execution; payouts, transfers, disputes, and application-fee refunds governed from webhooks. Only refunds routed through the enforcing path can be blocked before execution; webhooks observe actions after Stripe executes.

USDC on SolanaLive, enabled per organization

Stablecoin transfers initiated by treasury-management agents. Pre-execution transfer decisions are live via the public API, enabled per organization.

Four primitives

The contract for agent money movement.

Four primitives compose into the contract every governed rail honors. Build new rails on top, audit old rails against it, never blur the boundary.

One decision endpoint

Requests routed through Axiru are evaluated by the Decision Engine. Direct provider calls and webhook-only observation are not pre-execution enforcement. Policies match on rail, rail action, amount, initiator kind, agent scope, counterparty, rolling windows, and time of day, ten rule kinds in the v2 DSL.

Signed pre-authorizations

The x402 pre-authorization path returns a signed JWT only on allow. The integrated caller/verifier must reject a missing or invalid authorization before settlement. Stripe DAA remains shadow-only.

Tamper-evident audit ledger

Every decision lands in the audit ledger with the OVT fingerprint, rail, rail action, policy versions evaluated, agreement vs the v1 shadow result, and the signed authorization JTI. SHA-256 chained. Auditors re-derive end-to-end.

Shadow before enforcing

Evaluate policy without blocking first. Enable only a supported enforcing integration after validating the caller and verifier path; shadow observations alone cannot block payment.

How it works

From agent request to signed authorization in one round-trip.

The same flow ships across rails. Stripe DAA and x402 use the same authorization endpoint; the rail-specific bits live in the OVT and the verifying middleware. Your policies don't care which rail the request came in on, unless you want them to.

  1. Agent requests an authorization

    For x402, POST /api/v1/authorizations with challenge (payment requirements) and context (initiator and optional idempotency key). USDC on Solana uses its separate rail, transfer and context shape. Use the documented schema for your rail.

  2. Decision Engine evaluates

    Policies run in priority order. The v1 evaluator returns the canonical decision (allow / deny / require_approval). The v2 cross-rail evaluator runs in shadow and records agreement metrics. Both paths share the same Prisma policy rows.

  3. Signed authorization or escalation

    On allow, the x402 route returns a signed token bound to the proposed intent. Deny or require_approval returns no spend authorization. Signing or evidence-write failure also withholds the token.

  4. Integrated caller verifies before settlement

    An integrated x402 caller/verifier checks the authorization against the published JWKS before proceeding. Provider settlement is separate from policy authorization. This does not describe enforcing Stripe DAA, which is shadow-only today.

Demo

Explore a sample refund decision ledger.

The dashboard uses illustrative refund data. It shows policy outcomes and review workflows; it does not run an x402 handshake, verify a live token or settle a payment.

Authoring

Same policy editor, new rail-agnostic rule kinds.

The v2 DSL adds ten rail-agnostic rule kinds, rail, rail_action, amount, initiator_kind, initiator_id, agent_scope, counterparty, rolling_window, time_of_day, custom_expression. Your existing v1 policies keep evaluating exactly as today.

  • Rail filter: apply a policy only to Stripe DAA or x402 traffic, leave the legacy Stripe refund path untouched.
  • Rolling-window aggregates: match when any single agent has spent more than $5,000 in the rolling 24h window.
  • Counterparty country lists: evaluate destination-country rules in shadow mode. A recorded deny is not an executed DAA block.
  • Agent scope guards: require payments:write; refuse if dangerous:cross-tenant is present.
  • DAA stablecoin starter templates: three opinionated templates ship today: browse the library →
Next step

Start with a supported payment path.

Start with read-only replay. Validate the enforcing caller path before enabling a supported integration.

Start with a free replay. Turn on live enforcement later.

We use cookies for analytics and marketing measurement. You can reject non-essential cookies at any time.

Privacy policy