The dictionary is converted to a JSON by the jsonify() function and returned as a HTTP response. ReflectionSimple assigns some value to internal properties and returns a dictionary of properties in the ReflectionSimple.getDict() function. The ‘/Simple/’ route calls the function get_simple(data) which creates an object of type ReflectionSimple and using the value that is passed to the GET call. This route is not meant to be used by Copilot, so we will not focus on this. Note that if someone navigates to the main site (‘/’ route) they will get a HTML message as shown in the main(). The purpose of having the optional ‘/file/’ route will be given later. This webservice exposes two GET routes one for ‘/Simple/’ and another optional route ‘/file/’. # Specify Network interface to bind to and the port to run

Return send_file(filePath,download_name=fileName,as_attachment=True) #- NOTE: All files should be placed in the.

#- Use this URL to return YAML files if you do not want to the host the files in FTP/Azure/Blob

Text += ' - GET /simple/ - Will return a JSON contains Please use the following REST API routes to get specific JSON results.' Text = 'Welcome to Copilot for Security test REST APIs. # When the main page is visited via browswer this page will be main(): # Show the REST APIs supported and format as a basic HTML. #- Define classes that will be serializedĭef _init_(self,data,json,ip,useragent): The code of this webservice is given below:įrom flask import Flask, request, jsonify, send_file We will build this web service in Python using the Flask web framework. If you want to make REST call to your own web application or test http servers, feel free to skip this part and go to Section 2. You can also use some of the test http servers like or. While we can use any standard application that exposes a REST API, it will be easier and clearer from the server-side if we build a simple web service of our own and have our plugin call it. If you want to know more about the existing 3-P plugins that are currently available on Copilot please click here. Note that this article discusses building new plugins from scratch. The code for the Webservice and the YAML files is available under Workshops in the Copilot GitHub page here. This article assumes you have basic familiarity with the Python programming language and the YAML file format.

Since JSON (JavaScript Object Notation) is the most popular format used in REST APIs that is the output format we will be using for our plugins. In a follow-up article we will show how to build API plugins for more complex GET and POST calls which involve the passing of parameters. In this article, we how to build a simple API plugin that utilizes a GET request with no parameters. API plugins allow Copilot to make REST API calls, to both Microsoft and non-Microsoft services and hence are the most powerful and relatively complex of the plugins. At a high level, GPT plugins allow you to call specific skills inherent in the LLM, while KQL plugins execute Kusto Query Language (KQL) queries to pull data from Microsoft Defender 365 products like Sentinel and Defender. More details and the differences between these plugins are given here. Plugins in Copilot can be of three types: The plugin mechanism allows Copilot to not only pull data from Microsoft products but also from third-parties. Copilot is not a monolithic system, but is an ecosystem running on a platform that allows data requests from multiple sources using a unique plugin mechanism. Copilot for Security (Copilot ) is a large language model (LLM) based generative Artificial Intelligence (AI) system for cybersecurity use cases.

Download JSON Editor for windows