x402.professorsausages.com

{
  "properties": {
    "articles": {
      "items": {
        "properties": {
          "categories": {
            "items": {
              "type": "string"
            },
            "type": "array"
          },
          "cluster": {
            "description": "Story-cluster id — articles sharing one are the SAME story from different outlets. The served window already keeps only one article per cluster, so this is provenance, not a filter you must apply. null when the headline was too short to cluster safely.",
            "type": [
              "string",
              "null"
            ]
          },
          "country": {
            "type": [
              "string",
              "null"
            ]
          },
          "domain": {
            "type": "string"
          },
          "feed": {
            "description": "Human feed label, e.g. \"BBC World\", \"Hacker News\", \"GDELT: technology\".",
            "type": "string"
          },
          "id": {
            "description": "sha1(normalized url) hex, first 16 chars — the dedupe key.",
            "type": "string"
          },
          "lang": {
            "description": "ISO-639-1 language, best-effort.",
            "type": [
              "string",
              "null"
            ]
          },
          "published": {
            "description": "ISO-8601 publish time (best-effort).",
            "type": "string"
          },
          "seen_at": {
            "type": "string"
          },
          "snippet": {
            "description": "Short plain-text description from the feed (≤240 chars); absent when the feed carries none.",
            "type": [
              "string",
              "null"
            ]
          },
          "source": {
            "enum": [
              "gdelt",
              "hn",
              "rss",
              "gnews",
              "sitemap"
            ],
            "type": "string"
          },
          "tier": {
            "description": "Reputability tier of the feed: official (Fed/SEC/White House/BLS/ECB press), major (wires + major outlets), trade (quality trade press), social (Hacker News), enrichment (GDELT-discovered, reputation-gated). Absent on records archived before the tiered portfolio.",
            "enum": [
              "official",
              "major",
              "trade",
              "social",
              "enrichment",
              null
            ],
            "type": [
              "string",
              "null"
            ]
          },
          "title": {
            "type": "string"
          },
          "tone": {
            "description": "GDELT document tone (negative = adverse), attached by URL match from GDELT's 15-minute files — present on any source GDELT also saw; null otherwise.",
            "type": [
              "number",
              "null"
            ]
          },
          "url": {
            "type": "string"
          }
        },
        "required": [
          "id",
          "title",
          "url",
          "domain",
          "source",
          "feed",
          "published",
          "categories",
          "seen_at"
        ],
        "type": "object"
      },
      "type": "array"
    },
    "count": {
      "type": "integer"
    },
    "freshness": {
      "properties": {
        "as_of": {
          "description": "updated_at of the corpus this response was served from.",
          "type": "string"
        },
        "coverage": {
          "properties": {
            "articles": {
              "description": "Articles covering that span.",
              "type": "integer"
            },
            "hours": {
              "description": "newest − oldest in hours — the searchable depth, usually far below window_hours.",
              "type": "number"
            },
            "newest_published": {
              "description": "Newest publish time held.",
              "type": [
                "string",
                "null"
              ]
            },
            "oldest_published": {
              "description": "Oldest publish time held; the real floor of any date filter.",
              "type": [
                "string",
                "null"
              ]
            }
          },
          "required": [
            "oldest_published",
            "newest_published",
            "hours",
            "articles"
          ],
          "type": "object"
        },
        "source_count": {
          "description": "Number of sources with ≥1 article in the current pile.",
          "type": "integer"
        },
        "window_hours": {
          "description": "Retention POLICY in hours (72). NOT the searchable depth — see coverage.hours.",
          "type": "integer"
        }
      },
      "required": [
        "as_of",
        "window_hours",
        "source_count",
        "coverage"
      ],
      "type": "object"
    },
    "updated_at": {
      "type": "string"
    },
    "window_hours": {
      "type": "integer"
    }
  },
  "required": [
    "updated_at",
    "window_hours",
    "count",
    "articles",
    "freshness"
  ],
  "type": "object"
}