Interface between ChatGPT and ERP system to add article descriptions and SEO data
Budget: €250 – €750 EUR
There is currently the problem that the descriptions and SEO data are not maintained for many articles in the ERP system. A new interface between ChatGPT and the ERP system should now provide a remedy. As it has to be a Windows application, the programming language C# is to be used for this. It contains a clear, logically structured graphical user interface.
The desired functionality is as follows:
1. the interface reads all article data records from the ERP system via API where the short or normal description, as well as SEO data such as the title tag, meta keywords and the meta description do not contain a value. These data records to be revised must first be cached locally. The readout contains the internal key (primary key in the database), the article number and the article name, as well as information on whether a short or long description is available, or whether the SEO title tag, meta keywords or meta description are set.
2. now that all relevant data records are known, the interface will ask ChatGPT via API to complete the corresponding data record. In the graphical interface, the user has the option of choosing between the various AI models that can be mapped by the ChatGPT API. There must be a link indicating that the respective models incur different costs. The previously read article name is transferred to ChatGPT and the intermediate result obtained is initially stored locally. There are several filter options for each corresponding data set, such as a maximum character length for descriptions, title tag and meta description, or how many keywords should be used, separated by commas.
3. now that all data has been completed by ChatGPT, this result should be clearly presented to the user in a graphical interface so that he can view it for all completed data records. The data records that the user is satisfied with so far can be marked in the graphical interface and can then be transferred to the ERP system with a further click. The data is written to the ERP system via API. For the data records that were not satisfactory, ChatGPT can be triggered again for processing.
The desired functionality is as follows:
1. the interface reads all article data records from the ERP system via API where the short or normal description, as well as SEO data such as the title tag, meta keywords and the meta description do not contain a value. These data records to be revised must first be cached locally. The readout contains the internal key (primary key in the database), the article number and the article name, as well as information on whether a short or long description is available, or whether the SEO title tag, meta keywords or meta description are set.
2. now that all relevant data records are known, the interface will ask ChatGPT via API to complete the corresponding data record. In the graphical interface, the user has the option of choosing between the various AI models that can be mapped by the ChatGPT API. There must be a link indicating that the respective models incur different costs. The previously read article name is transferred to ChatGPT and the intermediate result obtained is initially stored locally. There are several filter options for each corresponding data set, such as a maximum character length for descriptions, title tag and meta description, or how many keywords should be used, separated by commas.
3. now that all data has been completed by ChatGPT, this result should be clearly presented to the user in a graphical interface so that he can view it for all completed data records. The data records that the user is satisfied with so far can be marked in the graphical interface and can then be transferred to the ERP system with a further click. The data is written to the ERP system via API. For the data records that were not satisfactory, ChatGPT can be triggered again for processing.