Appwrite Assistant is an AI-powered API that helps you with Appwrite-related tasks, powered by the official Appwrite documentation.
Make sure you have pnpm installed.
To install, run the following command.
pnpm i
Next, fetch the Appwrite-specific sources used by the assistant. This will download the sources from the Appwrite documentation and store them in the ./sources
directory.
pnpm run fetch-sources
The scripts will pull the latest documentation from the main
branch of the website repository, and the latest API reference from live Appwrite documentation.
If you want to pull from a different branch or repository, you can set the _BUILD_GIT_URL
and _BUILD_WEBSITE_URL
environment variables.
First, retrieve an API key from OpenAI. You can # for an API key at OpenAI. Once you have an API key, set it as the _APP_ASSISTANT_OPENAI_API_KEY
environment variable.
To run the server, execute the dev
command. By default, the server will be available at http://localhost:3000
pnpm run dev
The server exposes a POST endpoint at /
. The endpoint expects a raw text body containing the query for the assistant. The answer to the query will be streamed back to the client as raw text.
Use cURL to test the server, for example:
curl -X POST -H "Content-Type: application/json" -d "{\"prompt\": \"How do I create a new user?\"}" http://localhost:3000/v1/models/assistant/prompt
All code contributions, including those of people having commit access, must go through a pull request and be approved by a core developer before being merged. This is to ensure a proper review of all the code.
We truly ❤️ pull requests! If you wish to help, you can learn more about how you can contribute to this project in the contribution guide.