Skip to content

Latest commit

 

History

History
93 lines (68 loc) · 4.26 KB

README.md

File metadata and controls

93 lines (68 loc) · 4.26 KB
Raw

@ubiquibot/plugin-template

Prerequisites

  • A good understanding of how the kernel works and how to interact with it.
  • A basic understanding of the Ubiquibot configuration and how to define your plugin's settings.

Getting Started

  1. Create a new repository using this template.
  2. Clone the repository to your local machine.
  3. Install the dependencies preferably using yarn or bun.

Creating a new plugin

  • If your plugin is to be used as a slash command which should have faster response times as opposed to longer running GitHub action tasks, you should use the worker type.
  1. Ensure you understand and have setup the kernel.
  2. Update compute.yml with your plugin's name and update the id.
  3. Update context.ts with the events that your plugin will fire on.
  4. Update manifest.json with a proper description of your plugin.
  5. Update plugin-inputs.ts to match the with: settings in your org or repo level configuration.
  • Your plugin config should look similar to this:
plugins:
  - name: hello-world
    id: hello-world
    uses:
      - plugin: http://localhost:4000
        with:
          # Define configurable items here and the kernel will pass these to the plugin.
          configurableResponse: "Hello, is it me you are looking for?"
          customStringsUrl: "https://raw.githubusercontent.com/ubiquibot/plugin-template/development/strings.json"
At this stage, your plugin will fire on your defined events with the required settings passed in from the kernel. You can now start writing your plugin's logic.
  1. Start building your plugin by adding your logic to the plugin.ts file.

Testing a plugin

Worker Plugins

  • yarn/bun worker - to run the worker locally.
  • To trigger the worker, POST requests to http://localhost:4000/ with an event payload similar to:
await fetch("http://localhost:4000/", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    stateId: "",
    eventName: "",
    eventPayload: "",
    settings: "",
    ref: "",
    authToken: "",
  }),
});

A full example can be found here.

Deploying the Worker

For testing purposes, the worker can be deployed through the Worker Deploy and Worker Delete workflows. It requires to create a personal Cloudflare Account, and fill the CLOUDFLARE_ACCOUNT_ID and CLOUDFLARE_API_TOKEN within your GitHub Action Secrets.

Action Plugins

  • Ensure the kernel is running and listening for events.
  • Fire an event in/to the repo where the kernel is installed. This can be done in a number of ways, the easiest being via the GitHub UI or using the GitHub API, such as posting a comment, opening an issue, etc in the org/repo where the kernel is installed.
  • The kernel will process the event and dispatch it using the settings defined in your .ubiquibot-config.yml.
  • The compute.yml workflow will run and execute your plugin's logic.
  • You can view the logs in the Actions tab of your repo.

Nektos Act - a tool for running GitHub Actions locally.

More information

  • Full Ubiquibot Configuration - helpful for defining your plugin's settings as they are strongly typed and will be validated by the kernel.
  • Ubiquibot V1 - helpful for porting V1 functionality to V2, helper/utility functions, types, etc. Everything is based on the V1 codebase but with a more modular approach. When using V1 code, keep in mind that most all code will need refactored to work with the new V2 architecture.

Examples