Questions, answered.

Neptuv is a trust score for AI agents. It reads the public ERC-8004 registries on Ethereum, Base and other chains and turns what it finds into one score from 0 to 100, with the reasons written in plain words. Businesses use it to decide whether to let an agent buy from them, call their API or access their data.

A number from 0 to 100 built from four groups of public signals: Reputation (up to 35 points), Validation (up to 25), Identity (up to 20) and Track record (up to 20). Points are taken away for risk flags. 80–100 is Trusted, 60–79 Good, 40–59 Caution and 0–39 Risky.

ERC-8004 ("Trustless Agents") is a draft Ethereum standard that gives AI agents an on-chain identity. It defines three registries: Identity (each agent is an NFT pointing to a registration file), Reputation (feedback from clients) and Validation (results from independent checkers). The registries are deployed at the same addresses on many chains.

Neptuv adds up points from four signal groups and subtracts penalties. Feedback counts more when it comes from many different, established reviewers and when it is recent. Independent validations count more when several validators agree. The full method, with every weight, is published on the methodology page.

It means fewer than three different reviewers and validators have looked at the agent so far. Being new is not a red flag, so Neptuv shows "Not enough data" instead of calling a new agent risky.

No. Paid plans unlock tools such as the verified badge, alerts and API access. The score is calculated the same way for every agent, only from public data.

The indexer reads new blocks about every 10 minutes and re-scores any agent with new feedback, validations or ownership changes. Registration files and endpoints are re-checked weekly.

ERC-8004 already blocks owners from rating their own agents. On top of that, Neptuv weights each reviewer once no matter how many entries they post, gives brand-new wallets that rated only one agent 40% weight, gives reviewers who only ever rate one owner's agents 40% weight, ignores ratings traded between owners who review each other (and takes 10 points away), ignores feedback from the owner's other wallets, and takes 15 points away when most feedback arrives in a 24-hour burst from brand-new wallets. Each agent page shows a trust graph of who vouches for it.

No. Neptuv only reads public data. Subscription payments go straight from your wallet to Neptuv's receiving wallet, and Neptuv only checks the blockchain to confirm they arrived. Neptuv never takes custody of anyone's funds and never asks for a private key or seed phrase.

With USDC on Base, or USDC, USDT or ETH on Ethereum. Plans are prepaid for 1, 3 or 12 months, and a year costs the same as 10 months. Nothing is ever charged automatically; your dashboard, and Telegram or your webhook if you set them up, remind you a week before a plan ends.

Sign in with the wallet that owns the agent on the ERC-8004 Identity Registry. Signing in is a free signature and costs no gas, and the claim is verified instantly.

Yes. Owners can open a dispute from the agent's page. Every correction is logged publicly on that page, and the method is versioned so anyone can see how scores are made.

No. The score is information calculated from public data, with no guarantee. An agent with a high score can still fail, so use it as one input alongside your own checks.

Through the API, the SDKs or the MCP server: look up any agent's score and reasons, verify that a request really comes from the agent it claims to be (Know Your Agent), and get an allow or deny answer from your own saved policy, such as a minimum score or required validations. Every decision is logged for audits.

Know Your Agent is checking which AI agent is calling you and whether it can be trusted, before it buys, books or reads data. With Neptuv, the agent signs each request with the wallet registered for it on ERC-8004; Neptuv verifies the signature against the registry, blocks replays and applies your rules in one API call.

Paying Pro and Platform workspaces can report agents that scammed, spammed or failed to deliver. Reports stay private until Neptuv reviews them, and the owner can reply. A confirmed report adds a public risk flag (25 points) with Neptuv's own short summary, and every platform checking the agent sees it.

Yes. On the Platform plan Neptuv issues signed EIP-712 attestations of an agent's score. A small open-source Solidity contract verifies them on-chain, so a protocol can require a minimum trust score without calling an API or trusting an oracle.

Yes. Neptuv runs an MCP server, so Claude, ChatGPT, Cursor and custom agents can check, search and compare agents directly. There is also a Telegram bot for quick checks and alerts.