OCR (formerly ChatOCR) · GPT-Aktionen

{
  "openapi": "3.0.1",
  "info": {
    "title": "OCR Plugin",
    "description": "The best way to read text from any document. Extract text from scanned PDFs, photos, and even handwriting.",
    "version": "v1"
  },
  "servers": [
    {
      "url": "https://docit.ai"
    }
  ],
  "paths": {
    "/upload_url": {
      "post": {
        "operationId": "uploadUrl",
        "summary": "Upload a URL to a file and return the OCR data. After the first upload, the job_id and next_token should be provided in the request body to continue the request.",
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "type": "object",
                "properties": {
                  "url": {
                    "type": "string",
                    "description": "The URL of the file"
                  },
                  "job_id": {
                    "type": "string",
                    "description": "The OCR job ID"
                  },
                  "next_token": {
                    "type": "string",
                    "description": "The next token to use for paginated responses"
                  },
                  "has_more": {
                    "type": "boolean",
                    "description": "Indicates that ChatGPT should continue the request and re-provide the job_id and next_token."
                  }
                }
              }
            }
          }
        },
        "responses": {
          "200": {
            "description": "OK",
            "content": {
              "application/json": {
                "schema": {
                  "type": "object",
                  "properties": {
                    "content": {
                      "type": "string",
                      "description": "The OCR data for the file"
                    },
                    "job_id": {
                      "type": "string",
                      "description": "The OCR job ID",
                      "nullable": true
                    },
                    "next_token": {
                      "type": "string",
                      "description": "The next token to use for paginated responses",
                      "nullable": true
                    },
                    "has_more": {
                      "type": "boolean",
                      "description": "Whether more OCR data is available for pagination. When true, this indicates there should be another call to the /upload_url endpoint."
                    }
                  }
                }
              }
            }
          }
        }
      }
    },
    "/queryDoc": {
      "post": {
        "operationId": "queryDoc",
        "summary": "Query a previously uploaded document. The job_id and query should be included in the request.",
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "type": "object",
                "properties": {
                  "job_id": {
                    "type": "string",
                    "description": "The OCR job ID"
                  },
                  "query": {
                    "type": "string",
                    "description": "The query about the uploaded document"
                  }
                }
              }
            }
          }
        },
        "responses": {
          "200": {
            "description": "OK",
            "content": {
              "application/json": {
                "schema": {
                  "type": "object",
                  "properties": {
                    "content": {
                      "type": "string",
                      "description": "The queried data for the file"
                    }
                  }
                }
              }
            }
          }
        }
      }
    }
  }
}