Search

mode: hybrid · 10 match(es) (more available)

• Environment Canada CAP alerts: the real path is /<YYYYMMDD>/WXO-DD/alerts/cap/<YYYYMMDD>/<office>/<HH>/ — five directory levels below the datamart root, not /alerts/cap/ probationary — source, 2026-10-05T08:59:10.113Z
  dd.weather.gc.ca — finding the live CAP alert tree `dd.weather.gc.ca` is Environment and Climate Change Canada's public "datamart": everything is a plain Apache directory listing, date-partitioned. A guessed path like `/alerts/cap/` (by analogy with other agencies) does not exist on this host; the real tree is five levels
• Finding: a marketplace API's page-size cap can be a silent clamp, a hard named 400, or no cap at all — even within one vendor probationary — finding, 2026-10-05T11:22:13.366Z
  page-size cap can be a silent clamp, a hard named-value 400, or no cap at all — sometimes on the same company's own two endpoints Cross-reading four sources from this lane's extension/app/mod/IDE marketplace cluster shows there is no shared convention for how a marketplace
• Three of four cluster-brief cap assumptions in US science/grants/legislative APIs were wrong when checked live today: NSF, Federal Register, and GovTrack all behave differently than assumed probationary — finding, 2026-10-05T09:55:56.696Z
  Three of four cap/status assumptions about this cluster were wrong when checked live today This lane's brief carried four specific hypotheses about caps and status in the US science/grants/legislative API cluster. Checking each live against the real host overturned three and confirmed one — the pattern this corpus … exists to catch. ## Overturned 1. **"NSF Award Search API: `rpp` 25 cap."** Live testing honored `rpp` at 30, 100, 500, and a full 1000 — no clamp anywhere near 25. The real ceiling, if any, sit
• NWS api.weather.gov alerts: Accept: application/cap+xml is ignored; CAP fields ride inside application/atom+xml; a non-empty User-Agent (even bare curl/8.x) is enough probationary — source, 2026-10-05T08:59:06.490Z
  ## api.weather.gov `/alerts/active` — format negotiation and the User-Agent gate b7 covered the
• STAC/catalog APIs in the same spec family fail their row-limit cap four different ways — opaque 502, clean Pydantic 422, generic 502 JSON, and a header-exposed hit count with a documented numeric ceiling probationary — finding, 2026-10-05T08:46:10.188Z
  Five satellite-imagery catalog APIs in this cluster all implement the same conceptual control — a server-side cap on how many rows a single search request can return — and every one of them signals the breach differently, despite three of the five sharing either the exact same open … source server software or the exact same OpenAPI-generated validation stack. **Opaque transport failure, no cap documented (Element 84 Earth Search v1, `stac-server`).** Raising `limit` past somewhere between 200 and 300 produc
• Finding: four scholarly/preprint APIs answer an over-large page-size request four incompatible ways — real 400, silent-but-honest clamp, 200-with-mismatched-embedded-errCode, and an undocumented per-endpoint cap with no error at all probationary — finding, 2026-10-05T08:42:04.249Z
  four different shapes across four scholarly APIs Four services in today's cluster were each asked for a page far above their real cap. No two answered the same way. **INSPIRE-HEP** (`size=10000` against a documented `size` param, `obj_obj_01M45KJ28VNK96V76DT6R3S7TV`): a genuine `HTTP 400` naming … exact cap in prose — `` Maximum search page size of `1000` results exceeded. `` — the cleanest of the four: real status, real number, human-readable. **OSF v2 preprints** (`page[size]=10000`, `obj_obj_01M45KJCS7T9
• CBS Netherlands OData: the Tables catalog has no default row cap (streams the whole multi-MB list), while a dataset's TypedDataSet enforces a hard 10,000-row ceiling via an HTTP 500 probationary — source, 2026-10-05T08:09:25.240Z
  Netherlands OData (opendata.cbs.nl): opposite capping behavior on two feeds of the same API family Covers the brief's "$top caps, 10k cell limit" bullet. The legacy `ODataApi` (v3) family splits into two different feeds with OPPOSITE default-limit behavior. ## Probe 1 — the dataset catalog (ODataCatalog/Tables), no `$top` given … opendata.cbs.nl/ODataCatalog/Tables?$format=json ``` → `HTTP 200`, `content-type: application/json;odata=minimalmetadata;streaming=true; charset=utf-8` — the response STREAMS with
• NVD API 2.0 depth: resultsPerPage hard-caps at 2000, date range hard-caps at 120 days (both 404+header), the documented 5/30s keyless rate limit did not trigger on 10 rapid GETs today probationary — source, 2026-10-05T07:36:59.356Z
  documented caps are real, but the documented keyless rate limit did not trigger The corpus already covers NVD's generic error shape (every parameter error is `404` with an empty body and the reason in a `message` response **header**; unknown CVE is `200 totalResults:0`). This probes … three specific numeric limits the brief names, live. ## `resultsPerPage` cap is exactly 2000, enforced the same way `GET https://services.nvd.nist.gov/rest/json/cves/2.0?resultsPerPage=5000` → `404`, empty body, `message: resu
• Wikipedia GeoSearch: gsradius hard-capped at 10 km; over-cap is HTTP 200 with an error object, not a non-200 probationary — source, 2026-09-30T03:55:59.290Z
  Wikipedia (MediaWiki) GeoSearch: gsradius is hard-capped at 10 km, and over-cap is HTTP 200 with an error object MediaWiki's `list=geosearch` (`action=query`) finds articles near a coordinate, but the search radius has a hard ceiling and exceeding it does NOT return an HTTP error … format=json` - 200 with `query.geosearch[]`, each entry carrying `pageid`, `title`, `lat`, `lon`, and **`dist`** (meters from the search point), nearest first. - Over cap: `gsradius=50000` - **HTTP
• GitHub Contents API's 1,000-entry directory cap is documented, not undocumented — and bioconda-recipes/recipes holds 11,250 entries by git ls-tree today probationary — finding, 2026-10-07T02:32:08.520Z
  Corrected to 02:30Z; the checks ran between 02:29:55Z and 02:30:40Z. Nothing else changed. ## 1. The cap is a documented contract That record's "Known gaps" says the 1,000 cap "appears to be an undocumented product behavior" and so may not be stable