Recommend
Surface movement or allocation actions with context.
Product / TreasuryAI
AI-native treasury recommendations inside policy bounds.
TreasuryAI is the command center for recommendations, approvals, movement controls, and reporting across treasury wallets.
AI-native, controlled: Recommendations can be useful only when policy, risk, human approval, and reconciliation stay attached.
Product console
Recommendations should never bypass controls.
TreasuryAI is designed around bounded autonomy: useful treasury insight, visible approval state, and evidence that finance can trust.
Operating boundary
Treasury inputsBalancesLiquidityRiskPolicy
TreasuryAI command layerSignalsRecommendationsApprovalsReports
Execution boundaryWalletKitApproved providersHuman reviewers
Evidence outputsDecision trailMovement receiptFinance report
Control loop
AI proposes; Alloy records the authority chain.
Approve
Keep human and policy gates in the path.
Report
Tie outcomes back to finance evidence.
Constrain
Keep recommendations inside approved assets, providers, limits, and time windows.
Treasury states
Recommendations need operating states before they need autonomy.
Opportunity
Signal appears with balance, liquidity, provider, and policy context.
Proposed action
The recommendation names amount, asset, destination, purpose, and expected evidence.
Approved movement
Human and policy approval are recorded before execution.
Closed outcome
The result flows into reporting, reconciliation, and exception handling.
Evidence
Proof the buyer can inspect.
Every route keeps Alloy's public story grounded in operating records, custody boundaries, and right-sized modules.
Next step
Bring your current provider stack.
We will map the first workflow Alloy should stabilize, the custody boundary, and the modules that create evidence.
Join design partners
Prefer email? hello@alloy.build