Frequently asked questions

An independent service that publishes an assessment of a DeFi protocol’s risk — a rating, a research report, or a live monitoring signal. The 16feeds in this registry range from formal rating methodologies to security-monitoring tools and research desks. Each looks at risk through its own lens. You can see the full list, with each provider’s focus, on the feeds page.

No — they differ in focus (some weigh smart-contract risk, others governance, liquidity, or market risk), in methodology, and in how much of the protocol landscape they cover. That is precisely why we put them side by side rather than picking a winner: no single feed is canonical for something this important. We show what each one says, attributed and verbatim, and let you weigh them. We never rank the feeds or merge them into a score of our own.

Because the aggregation is the value. The information already exists across dozens of dashboards and reports; what is missing is a neutral layer that puts it in one place. We treat risk feeds the way Ethereum treats price oracles — diversity is the point. Adding our own composite score would just create one more opinion to reconcile, and it would compromise our neutrality. This constraint is binding; see the charter.

Coverage describes whether a feed has published an assessment of a given protocol, not how risky the protocol is. Covered means the feed reports on the protocol; partial means limited or indirect coverage; and not yet covered means the feed has not published on it. A protocol with fewer feeds reporting on it has thinner coverage — a fact about the feed ecosystem, not a verdict on the protocol.

Where a feed publishes a rating or summary, we show it as the feed states it, attributed to the provider, with a link to the source — we do not paraphrase it into a judgement of our own. In this prototype some ratings are illustrative sample values pending provider verification, and are labelled as such.

We show the disagreement. Divergence between feeds is signal, not noise — it tells you risk on that protocol is contested and worth a closer look. We do not adjudicate, average, or pick a side.

By design, and by binding commitment. Producing a composite or proprietary risk score — or ranking protocols by any risk judgement of our own — is exactly what this project does notdo. We sort and filter only by neutral, externally-sourced facts (TVL, category, feed-coverage count). The full list of what we do and don’t do is on the methodology page; changing the constraint requires written agreement from the Ethereum Foundation.

A ratings agency produces its own verdict. We produce none — we show what others publish, side by side, and add a verification layer so you can check those claims against what actually happened on-chain. We are an aggregator and a verifier, not a rater.

The on-chain facts and the transaction-transparency links are produced by Sieve, an open-source Ethereum indexer that syncs directly over the P2P network — no RPC provider, no API keys. Because the indexer is open and re-runs over public data, every on-chain figure here is independently reproducible. The feeds tell you what they say; the verification layer shows what is.

No. Rails stewards and verifies the registry but is deliberately not a rated feed in the matrix — a steward grading itself alongside the feeds would not be neutral. Rails is infrastructure (provenance and on-chain verification) and earns no rating column. The relationship is disclosed in DISCLOSURES.md.

The registry seeds 20 protocols across 26 versioned deployments, chosen as top Ethereum protocols by funds at risk and curated for coverage across every category, and 16 feeds chosen to maximise methodology diversity. Selection uses only neutral, externally-sourced facts — never a risk judgement of our own. The full selection criteria are on the methodology page.

Yes — every datum is community-correctable. To fix a coverage status, a stale rating, or a governance detail, or to propose a new protocol or feed, open an issue or PR on GitHub citing the source. Maintainers verify against the cited source before merge.