This is a sequencing proposal, not a release promise or a list of implemented APIs.

Keep the Octant scope on incoming payments: signed request, usable checkout, separate payer execution, view-only reconciliation and deterministic merchant events. The v0.1.0-beta.3 release adds PayIn UX to the beta.2 baseline. Validate it with external adopters before broadening scope.

• Independently reproduce merchant onboarding and payer completion.
• Validate a compatible wallet handoff and improve reference-payer review/confirmation.
• Show actual network fee previews; estimate private-balance readiness only when reliable underlying data exists.
• Improve payer recovery and operational documentation from observed failures.

Balance preparation, swaps/funding, native wallet integrations and notifications need separate designs and user validation. They are not prerequisites added to the current merchant daemon.

Only after PayIn adoption and hardening, consider outgoing payment intents:

PPOps intent → external policy / approval → external spending wallet
             → private transfer → settlement monitoring → merchant event

PPOps coordinates and observes. A separately controlled wallet/signing service authorizes spending; the merchant daemon must not acquire spending keys or custody. Approval policies, recipient validation, idempotency, transaction recovery, fees, limits and optional multisig require a dedicated design and review first.

Refunds are the initial candidate because they relate directly to an accepted PayIn. Payroll, treasury and B2B disbursements are possible later uses, not current capabilities. POST /v1/payouts, payout states and payout webhooks do not exist.