Need help for some tools

[MrBrabus75]

Self Checks

• [X] I have searched for existing issues search for existing issues, including closed ones.
• [X] I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [X] Pleas do not modify this template :) and fill in all the required fields.

Provide a description of the help you need

Hello,

This is more of a question before making a request, is it possible to include a system like the OpenAI GPT-4 code interpreter with this: "https://github.com/olxver/sandbox-plugin/blob/main/versions/source/ai-plugin.json"? If yes, how? Will you be doing it, or should I do it myself?

Also, I have been trying for some time to add a parsing tool for Redmine and an internal site.

That is, having a Redmine tool that, when I ask a question about a ticket, it searches for the most relevant and likely tickets, gives me a summarized answer, and provides me with the link to the Redmine ticket(s). I can't seem to make it work, so I would need help if possible...

Additionally, I've seen that there is DuckDuckGo, etc., but I would like to know how to give it "https://mysite.fr" and have it parse the site to retrieve information, collect it, and respond to the question.

Any help is welcome. Thank you for everything.

Sincerely,

[dosubot[bot]]

Hey there, @MrBrabus75! Great to see you back and diving into some intriguing challenges. Looking forward to tackling these questions with you.

Based on the current functionality of the repository, it seems possible to integrate an OpenAI GPT-4 code interpreter as a plugin, provided it adheres to the ChatGPT Plugins standard. However, the specific process for integrating a new plugin into the Dify repository and the specific requirements for a plugin to be compatible with the Dify system are not explicitly documented in the repository.

For your second question, the repository currently does not have a built-in tool for parsing Redmine tickets. However, you could potentially develop a custom plugin that uses the Redmine API to search for tickets and provide summarized answers. This would involve using the Redmine API to fetch ticket data, processing the data to find the most relevant tickets, and then summarizing the ticket information.

As for parsing a specific site like "https://mysite.fr", you could potentially use a web scraping tool or library to retrieve information from the site. However, this would depend on the structure of the site and the specific information you want to retrieve.

I hope this helps! If you have any more questions or need further clarification, feel free to ask.

Sources

• en/features/datasets/README.md
• en/features/ai-plugins/README.md

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[Yeuoly]

I guess what you need is a customized tool, it searches things and summarize them, as a generic solution, Code Interpreter is a good choice, however, it causes high resource usage and brings security issues, that's why we could not provide a first party service of Code Interpreter. If you are using our SaaS, deploy this sandbox and create a custom tool of it will make sense, otherwise, make a RedMine tool by following our Tool Integration Guide could also work for you.