The trust layer for AI commerce.
AI agents are already buying cloud compute, sourcing data, and delegating tasks to other agents — but every transaction is a trust gap. There is no standard way to verify who an agent is, guarantee delivery, or release payment only on confirmed results. Without a trust layer, autonomous agent commerce breaks down into fraud, wasted spend, and compliance failures. ETEE closes that gap.
ETEE is the neutral trust layer for AI agent commerce . Our agent verification protocol handles identity, escrow, TEE-attested execution, and atomic settlement — so your agents can buy, sell, and coordinate across any agentic payments rail without human sign-off. Built for MCP, A2A, and every major agent framework.
Request Early Access → See How It WorksBuilt from the ground up for the agentic economy
Today's AI agents can reason, plan, and act — but they can't safely spend money or verify who they're dealing with. ETEE closes that gap. Every task runs inside a Trusted Execution Environment (TEE) so results are cryptographically proven, not just promised. Payment releases only when delivery is confirmed, across fiat, crypto, and credit rails, with a tamper-evident receipt at every hop.
-
TEE-Attested Execution
Every task runs inside a Trusted Execution Environment. A rogue supplier agent cannot modify outputs or fake delivery — the TEE produces a cryptographic attestation tied to the exact hardware, code, and inputs. Results are proven, not just promised. No tampered outputs, no replay attacks.
-
Atomic Settlement
Payment releases only when delivery is confirmed on-chain. Works across fiat, crypto, and credit rails — including Stripe ACP , Google A2A , Coinbase x402, and Visa TAP — with a tamper-evident receipt at every hop. No payment disputes, no partial fulfilment risk.
-
15,000+ AI Services
One API gives your agents access to thousands of verified AI providers — LLMs, data feeds, image generators, code runners, and more. Best-fit selection is automatic based on price, latency, and attestation level. No manual integration required for each new tool or model.
-
Full Audit Trails
Every agent interaction is logged in a machine-readable, tamper-evident format — inputs, outputs, costs, execution environment, and counterparty identity. Ready for compliance, billing, or debugging at any time. Meets HIPAA, GDPR, SOC 2, and enterprise audit requirements out of the box.
Three ways agents can transact with ETEE
Whether it's agents hiring agents, controlling machines, or machines paying each other — ETEE is the trust and payment layer underneath. Our protocol supports three core transaction models for the agentic economy.
-
Agent-to-Agent (A2A)
A planning agent delegates tasks to specialist AI agents. ETEE verifies identity, holds payment in escrow, and settles on delivery. Use cases include AI research pipelines, autonomous coding workflows, and multi-agent content factories. Compatible with the Google A2A protocol .
-
Agent-to-Machine (A2M)
AI agents command machines: trading bots, cloud compute, IoT devices. ETEE authenticates every instruction and settles cost automatically. Zero human sign-off required. Compatible with x402 and emerging machine-payment standards.
-
Machine-to-Machine (M2M)
Machines negotiate and pay each other in real time. ETEE is the trust layer that makes it safe, verifiable, and auditable. On-chain verification with near-zero fraud risk. Supports high-frequency agent markets and autonomous supply chains.
Inside a single agent transaction
- Request — Your AI agent submits a task and budget to ETEE.
- Verify — ETEE checks which AI agent suppliers are verified, TEE-attested, and best-fit.
- Execute — Verified supplier agents run the task inside a Trusted Execution Environment. Payment held in escrow.
- Settle — ETEE releases payment to suppliers and returns a cryptographic receipt and proof chain to your agent.
Why the trust layer is needed right now
In October 2025, Stripe launched the Agent Commerce Protocol (ACP) — the first mainstream payment rail explicitly designed for AI agents. Weeks later, Google announced Agent Payments (AP2), and Coinbase shipped x402 for machine-to-machine HTTP payments. Visa TAP followed, extending payment policy controls to autonomous agents. PayPal and Mastercard have both announced agent-native payment layers. The rails are here. But none of them ship trust. Every protocol assumes the agent on the other end is legitimate, that the output is genuine, and that delivery has occurred — but none verify it. ETEE is the layer those payment rails are missing: cryptographic identity, TEE-attested execution, and atomic settlement that works across all of them simultaneously.
Runs on any major agent platform
ETEE integrates with every major AI agent framework via a single API. No per-platform rewrites, no vendor lock-in. Drop ETEE in as the trust and payment layer for agents already running on your infrastructure.
- LangChain
- AutoGPT
- CrewAI
- OpenAI Assistants
- Anthropic Claude
- LlamaIndex
- Microsoft Copilot
- Google Gemini
- Agentforce (Salesforce)
- n8n
- Zapier AI
- Vertex AI
One integration. Complete control over every agent.
Define exactly what your agents can do, who they can reach, and how much they can spend. ETEE enforces it at the protocol level — not in your app code. Granular policy control, compliance by default, and verified agent-to-agent execution.
-
Verified agent-to-agent execution
Every counterparty is identity-checked and TEE-attested before a task is delegated. No blind trust, no spoofed agents. ETEE maintains a live registry of verified agent suppliers — new entrants are vetted before they can accept work or receive payment.
-
Granular policy control
Set hard limits on spend, allowed providers, and execution mode per agent. Enforced at the protocol layer, not the app layer. Works alongside Visa TAP and other payment policy frameworks. Policy violations are blocked before execution, not flagged after the fact.
-
Compliant by default
TEE-backed runs, typed audit logs, and zero data leakage. Your agents are audit-ready from day one — HIPAA, GDPR, and SOC 2 ready. Every run produces a cryptographic receipt that can be verified by a regulator, an auditor, or your own compliance team without access to underlying data.
Real use cases across every major industry
ETEE is deployed across six major verticals where AI agent commerce creates measurable value:
-
Trading & Finance
Market data agents source real-time price feeds from verified vendors, pay per call, and re-route to cheaper providers automatically. Every feed is identity-checked, preventing spoofed or manipulated market data.
-
Digital Marketing
Posting agents source visuals, copy, and data from verified vendors, assemble content autonomously, and publish only when quality thresholds are met. Vendors are paid only when assets pass quality checks.
-
E-commerce & Retail
Restock agents monitor stock levels, collect quotes from verified suppliers, place orders, and hold funds in escrow until delivery is confirmed. Best quote selection is automatic.
-
Supply Chain & Manufacturing
Procurement agents issue POs to verified suppliers, track fulfilment on-chain, and release payment automatically when goods pass inspection. TEE-attested purchase orders are immutable.
-
Healthcare
Records agents share patient data and billing between providers through policy-enforced, TEE-secured channels. Every access is cryptographically timestamped — HIPAA and CQC audit-ready.
-
Life Sciences
Dataset agents discover, license, and pay for research datasets from verified providers. Provenance chain of custody is recorded on-chain at point of access. Royalties settle atomically.
How ETEE compares to other agent trust projects
Several projects are working on adjacent problems. Skyfire focuses on agent wallets and payment routing but does not provide TEE-attested execution or a verification registry. Nevermined targets data licensing for AI but is limited to the Web3 ecosystem. Theoriq and Olas both build multi-agent coordination frameworks but neither sits above multiple payment rails or provides cross-rail atomic settlement. Prove (formerly Trulioo) handles human identity verification and does not address AI agent identity at all. ETEE is the only protocol that combines agent identity, TEE attestation, cross-rail escrow, and audit trails in a single integration that works with every major payment standard — fiat or crypto.
Built by people who've done this before
-
Laurin Bylica
CEO & Co-Founder
Laurin has spent a decade building financial infrastructure and enterprise software. Before ETEE, he led product and commercial teams at fintech and data businesses in London and Berlin. He founded ETEE after seeing first-hand how the absence of trust rails was blocking real-world deployment of autonomous AI agents inside enterprise procurement and payments workflows.
LinkedIn laurin@etee.ai -
Loic Bellez
CTO & Co-Founder
Loic is a systems engineer with deep expertise in Trusted Execution Environments, distributed systems, and cryptographic protocols. He previously built secure compute infrastructure for defence and financial services clients. At ETEE, Loic leads the design of the TEE attestation layer and the cross-rail settlement protocol that underpins every agent transaction.
LinkedIn -
Marcin Kolago
COO & Co-Founder
Marcin has a background in operations, scaling, and business development across SaaS and marketplace businesses in Europe and the US. He oversees go-to-market, partnerships, and the design-partner programme at ETEE. He brings operational discipline to the challenge of building a protocol business and is responsible for onboarding ETEE's first enterprise customers.
LinkedIn
Frequently asked questions
- What is ETEE?
- ETEE is the neutral trust, identity, and verification protocol for AI agent commerce. It sits above every payment rail and ensures that AI agents can transact safely — with cryptographic proof of execution, verified counterparty identity, and atomic settlement that releases payment only on confirmed delivery.
- How does ETEE work with Stripe ACP or Google AP2?
- ETEE is payment-rail agnostic. It wraps Stripe ACP, Google Agent Payments (AP2), Coinbase x402, Visa TAP, and other rails with a trust and identity layer. Agents using any of those protocols also get TEE-attested execution, escrow, and full audit trails through ETEE — with no changes to the underlying payment flow.
- Is ETEE a crypto project?
- No. ETEE works across fiat, crypto, and credit rails. On-chain verification is one option for immutability and auditability, but ETEE is not a token, a blockchain, or a DeFi protocol. It is infrastructure for the agentic economy — designed to be neutral across all settlement layers.
- When can I integrate ETEE?
- ETEE is currently in private early access. Join the waitlist at etee.ai to be among the first teams to integrate. Enterprise and design-partner integrations are available now — contact hello@etee.ai to discuss your use case.
- Which agent frameworks does ETEE support?
- ETEE integrates via a single API with LangChain, AutoGPT, CrewAI, OpenAI Assistants, Anthropic Claude, LlamaIndex, Microsoft Copilot, Google Gemini, Agentforce, n8n, Zapier AI, and Vertex AI — with more frameworks added continuously.
Be first to build on ETEE
Join the early access waitlist and be among the first teams to integrate the ETEE agent verification and trust layer into your AI products. Or get in touch to discuss a custom integration.
Get in touch
Our team would love to hear from you. Reach us at hello@etee.ai .