AI-agent data is fragmented across registries, chains, and protocol directories. Trustprint turns that sprawl into daily validated snapshots, public Trust Checks, and a clearer picture of which agent identities are real, active, and comparable.
Every registry has its own schema, its own API, and its own view of the world. The same agent registers on three platforms — and becomes three separate, unconnected entries. Builders can't compare. Researchers can't count. Investors can't size the market.
The agent economy runs on several protocol layers. Most tools see only one. Trustprint indexes the verifiable pieces first and keeps roadmap layers separate until their contracts are proven.
Today, Trustprint's public truth is the indexed agent dataset, daily snapshots, registry health, and the live v1.1 trust lookup. MCP directories, A2A/ACP cards, x402 activity, and richer identity links are tracked as distinct lanes when their public contracts are reproducible.
At the foundation, ERC-8004 gives many agents an on-chain registration footprint. Trustprint compares that footprint with other registries and evidence sources, but does not pretend every record is already resolved to a single verified operator.
The result: a disciplined intelligence layer that tells you what is verified today and what still needs stronger evidence.
Our four-stage pipeline resolves the same agent across registries — even when names, addresses, and schemas don't match.
It starts with wallet matching — if two entries share an ETH address, that's a near-certain link. Exact matching then compares names, frameworks, and categories against a normalized schema. For harder cases, proprietary fuzzy matching handles what rule-based systems miss. Edge cases go through an LLM confirmation step for final verification.
ERC-8004 gives an agent a fingerprint today. LUKSO Universal Profiles are the clearest future path to a portable operator identity that can link the same agent across chains and registries without guesswork. That makes LUKSO strategically important for Trustprint, but it remains an optional identity bridge rather than a live indexed source today.
What we publish publicly today is the verified layer: daily crawl health, validated snapshots, and the live `v1.1` Trust Check.
Five registry-specific scout roles run inside the daily crawl with probeability-aware quotas: successful checks and contract-reachable endpoints get more budget, while no_endpoint sources stay visible as small sentinel samples.
no_endpointEvery Scout interaction lands in scout_interactions with endpoint, latency, status, and error context. That gives Trustprint an operational coverage layer no static registry export can provide, but only for the sources that are actually probeable today. The moat is disciplined truthfulness about live reachability, not invented traffic.
Every agent interaction delegates trust to a counterparty you may know little about. A weak evidence profile is not a final verdict; the public v1.1 score is a review signal for indexed agents, while richer policy and identity-confidence layers come only after the evidence model is proven.
Enter any indexed EVM agent address to see the same live signal builders can use before trust, payment, or integration decisions.
Start with the live lookup, then move into the API contract and waitlist only if the signal is useful for a real workflow.
Use the free lookup now. Join the waitlist for Pro and Team access until paid plans are live.
All plans include the weekly Agent Economy Pulse newsletter. Paid access is currently waitlist-only until checkout is live.
What builders, researchers, and investors ask about the agent economy and Trustprint's role in it.
The weekly data signal from inside the agent economy. Growth numbers, registry shifts, protocol adoption, and what the on-chain data actually means — distilled for builders, researchers, and investors. Every Tuesday.