The governed execution layer for agentic workflows.
Your agents are spending money and taking actions autonomously. Attestify is how you prove every one of those actions was authorised, within policy, and auditable — with a signed receipt returned on every run.
Every call passes through the governed Router — it authenticates, routes, enforces spend limits, settles payment in USDC on Base, and returns a cryptographic receipt before your orchestrator sees the result.
Plans from $59/mo · Cancel anytime · API key in under a minute
The question every team should be asking
“When our agent spends money, takes an action, or makes a decision autonomously — how do we prove it was authorised, within policy, and auditable?”
Attestify is the answer. Every agent run passes through the Router — which enforces policy, settles payment, and returns a signed receipt with verification, evidence, and settlement metadata. Every paid action is governed, inspectable, and provable.
Works with your stack
What Attestify OS gives you
Every execution returns a signed receipt — agent ID, USDC cost, SHA-256 output hash, and Base settlement tx. Ready for audit, dispute, or compliance review the moment the run completes.
Context persists across runs so agents operate with continuity. No re-explaining the task, no duplicate work — just durable, stateful execution that picks up where it left off.
Set daily, monthly, and per-run USDC limits per agent in the Control Plane. Runs that exceed policy are blocked before they execute — not discovered after the invoice arrives.
Choose the correct agent for the task — analyst, coder, researcher, support — and call POST /api/run. Plain HTTPS, no SDK required. The router handles auth, pricing, and settlement automatically.
How it works
A builder or agent sends a task into Attestify.
The lane is selected — usually by setting agent_id directly.
The paid execution lane runs the task.
Loop ID, price, SHA-256 output hash, and tx metadata are returned.
Teams inspect activity, receipts, and outcomes after every run.
Control Tower
Governance built into every run.
The Control Tower is the policy layer above the Router. Set constraints once; every agent call is evaluated against them before execution begins.
- ✓Policy rules and spend limits per agent
- ✓Intent-based execution for sensitive actions
- ✓Human approvals for high-risk workflows
- ✓Scoped identity and consent boundaries
- ✓Financial Control Plane — daily, monthly, and per-run limits
- ✓Receipts and evidence for disputes, audit, and compliance
Where to go
Choose a subscription plan — Builder or Team — and get your API key via Stripe checkout. Start here.
Already subscribed? Provision your tenant API key in under a minute.
Best first stop if you want a real paid run in minutes. Covers wallet setup, routing, and receipts.
Understand when to choose a lane directly and when to ask Router first.
Compare lanes, prices, and when each one should or should not be used.
Inspect recent loops, receipts, paid activity, and system evidence.
Set spend limits, approval thresholds, and governance policies per agent — directly in your browser.
Read the technical model, endpoints, examples, and machine-readable surfaces.