{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://awesomecua.github.io/schemas/github-candidate-observation.schema.json",
  "title": "AwesomeCUA public GitHub candidate observation",
  "description": "Public GitHub metadata observations only. Every row remains candidate/relevance_pending; HTTP success, a topic tag, or a repository name does not establish CUA relevance, release completeness, licensing, or editorial approval.",
  "type": "object",
  "required": [
    "id",
    "repository_id",
    "canonical_id",
    "full_name",
    "url",
    "source_description",
    "github_topics",
    "private",
    "visibility",
    "fork",
    "archived",
    "created_at",
    "updated_at",
    "pushed_at",
    "metadata_status",
    "review_status",
    "scope",
    "first_observed_at",
    "last_observed_at",
    "known_names",
    "evidence",
    "metadata_receipt_id",
    "omitted_fields",
    "record_sha256"
  ],
  "properties": {
    "id": {
      "type": "string",
      "pattern": "^github-repository:[1-9][0-9]*$"
    },
    "repository_id": {
      "type": "integer",
      "minimum": 1
    },
    "canonical_id": {
      "type": "string",
      "pattern": "^github:[a-z0-9][a-z0-9-]*/[a-z0-9_.-]+$"
    },
    "full_name": {
      "type": "string",
      "pattern": "[A-Za-z0-9][A-Za-z0-9-]*/[A-Za-z0-9_.-]+$"
    },
    "url": {
      "type": "string",
      "format": "uri",
      "pattern": "^https://github\\.com/"
    },
    "source_description": {
      "type": [
        "string",
        "null"
      ],
      "maxLength": 2000
    },
    "github_topics": {
      "type": "array",
      "items": {
        "type": "string",
        "pattern": "^[a-z0-9][a-z0-9-]{0,49}$"
      },
      "uniqueItems": true,
      "maxItems": 100
    },
    "private": {
      "const": false
    },
    "visibility": {
      "const": "public"
    },
    "fork": {
      "const": false
    },
    "archived": {
      "type": "boolean"
    },
    "created_at": {
      "type": "string",
      "format": "date-time"
    },
    "updated_at": {
      "type": "string",
      "format": "date-time"
    },
    "pushed_at": {
      "anyOf": [
        {
          "type": "string",
          "format": "date-time"
        },
        {
          "type": "null"
        }
      ]
    },
    "metadata_status": {
      "const": "primary_metadata_observed"
    },
    "review_status": {
      "const": "candidate"
    },
    "scope": {
      "const": "relevance_pending"
    },
    "first_observed_at": {
      "type": "string",
      "format": "date-time"
    },
    "last_observed_at": {
      "type": "string",
      "format": "date-time"
    },
    "known_names": {
      "type": "array",
      "items": {
        "type": "string",
        "pattern": "[A-Za-z0-9][A-Za-z0-9-]*/[A-Za-z0-9_.-]+$"
      },
      "minItems": 1,
      "uniqueItems": true
    },
    "evidence": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "receipt_id",
          "run_id",
          "query_id",
          "position",
          "url",
          "request_url_sha256",
          "observed_at",
          "retrieved_at",
          "sha256",
          "bytes"
        ],
        "properties": {
          "receipt_id": {
            "type": "string",
            "pattern": "^receipt:[0-9a-f]{64}$"
          },
          "run_id": {
            "type": "string",
            "pattern": "^run:[0-9a-f]{24}$"
          },
          "query_id": {
            "type": "string",
            "pattern": "^(search:[0-9a-f]{24}|existing:[0-9a-f]{64})$"
          },
          "position": {
            "type": "integer",
            "minimum": 0
          },
          "url": {
            "type": "string",
            "format": "uri",
            "pattern": "^https://api\\.github\\.com/"
          },
          "request_url_sha256": {
            "type": "string",
            "pattern": "^[0-9a-f]{64}$"
          },
          "observed_at": {
            "type": "string",
            "format": "date-time"
          },
          "retrieved_at": {
            "type": "string",
            "format": "date-time"
          },
          "sha256": {
            "type": "string",
            "pattern": "^[0-9a-f]{64}$"
          },
          "bytes": {
            "type": "integer",
            "minimum": 0
          }
        },
        "additionalProperties": false
      },
      "minItems": 1,
      "uniqueItems": true
    },
    "metadata_receipt_id": {
      "type": "string",
      "pattern": "^receipt:[0-9a-f]{64}$"
    },
    "omitted_fields": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "field",
          "reason"
        ],
        "properties": {
          "field": {
            "const": "source_description"
          },
          "reason": {
            "enum": [
              "credential_shaped_string",
              "private_home_path",
              "url_userinfo"
            ]
          }
        },
        "additionalProperties": false
      },
      "uniqueItems": true
    },
    "record_sha256": {
      "type": "string",
      "pattern": "^[0-9a-f]{64}$"
    }
  },
  "additionalProperties": false
}