How well we know whether declared endpoints answer
endpoint_coverageBEFORE YOU TREAT ANY ENDPOINT AS RELIABLE, ask how well we know. There is no uptime number here and there will not be one, because the states mean different things and summing them produces a figure about nothing: a 402 is an x402 seller doing exactly what it advertises, a 401 or 403 is credentials WE do not hold, a 3xx is a move we chose not to follow, and a timeout describes OUR attempt rather than the host. Only the states marked fault: true are the endpoint failing. TWO POPULATIONS, each on its own denominator and NEVER added together: the HOSTS registered identities declare, probed daily under a deadline that sometimes expires before every host is reached — so a host we did not ask keeps whatever an earlier day found, and notAsked says how many — and the x402 RESOURCES sellers list, where a state is published only after three knocks because one failure is not a finding. The second number is the uncomfortable one and it is published anyway: most listed resources have never been knocked on at all, which is our reach and not a judgement of the sellers. Free, because a coverage claim nobody can audit is marketing rather than evidence.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| meta | No | How current this answer is and what it was read from: `freshness` and the ages behind it. Present on every answer from this server, whatever the tool. | |
| says | Yes | Both populations in prose, each with its own denominator, and what the fault count does and does not include. | |
| hosts | Yes | The daily HEAD probe of hosts registered identities declare. `declared`, `asked` and `notAsked` — the probe runs under a deadline, so these differ and the difference is hosts whose state is a memory. `lastRunAt`, `ageHours` and `stale` against a daily cadence. `staleHosts` counts hosts nobody confirmed inside the freshness window. `byState` is one entry per state with `hosts`, `agents` (identities standing behind them — one dead host can carry hundreds), `share`, `fault` (true, false, or NULL where the state is about our attempt or our credentials rather than the endpoint) and `says`. `agentsBehindFaults` counts identities behind a fault state only, built from the states that ARE faults rather than by negating `answering`. | |
| warnings | No | Anything the server wants a caller to know about THIS answer before acting on it. Empty is the normal case and is not a guarantee. | |
| cannotSee | Yes | What no figure here can reach: endpoints nobody declares, whether the declaring identity controls the host, whether a 402 offer can actually be signed (that needs the response BODY and comes from the knock flow), and whether an endpoint that answered answered anything good. | |
| resources | Yes | The x402 resources sellers list, on its OWN denominator. `declared`, `everKnocked` (knocked even once) and `confirmable` (knocked three times or more, the only ones whose state we will publish), with `everKnockedShare` and `confirmableShare` beside them, plus `knockedLastWeek` and `oldestKnockHours`. Read `confirmableShare` as the honest answer to "can ASPERN tell me whether an x402 endpoint works": today it is small. | |
| measuredAt | Yes | When these figures were counted. The host states come from the last probe RUN, whose own age is on `hosts.ageHours`; a fresh read of a stale probe is not a fresh fact. |