Docker Labs: GenAI | No. 6

Announcing AI Tools for Developers in Docker Desktop

We’ve released a simple way to run AI Tools in Docker Desktop. With the Labs AI Tools for Devs Docker Desktop extension, people who have been asking for a simple way to run prompts can easily get started. If you’re a prompt author, this also allows you to build, run, and share your prompts easier. Here’s how you can get started.

Get the Extension

https://hub.docker.com/extensions/docker/labs-ai-tools-for-devs

Once installed, enter an OpenAI key.

Inputting Project

With our approach, the necessary information a prompt needs should be extractable from a project. Add projects to run prompts against, then select a project to use.

Inputting Prompts

A prompt can be a git ref or a git URL (which will convert to a ref). For prompt authors, you can also add a local folder here. For quickstart, you can use https://github.com/docker/labs-ai-tools-for-devs/tree/main/prompts/poem as an example prompt.

Writing & Testing Your Own Prompt

Wrap your own tool with a prompt

1. Scaffolding a prompt folder: Copy an existing prompt from https://github.com/docker/labs-ai-tools-for-devs/tree/main/prompts, or create a folder containing README.md, 010_system_prompt.md, 020_user_prompt.md

2. Populate the user prompt and system prompt files with your own prompt content

010_system_prompt.md

020_user_prompt.md

3. To populate the variables in handlebar templates like `{{project-facts}}`, you can use an extractor or build your own.

4. Add tools in the README front-matter. Existing tools are registered in https://github.com/docker/labs-ai-tools-for-devs/blob/main/functions/registry.yaml, but you can also define your own in the prompt frontmatter:

Test

1. Add the folder to a Git repository (not the root) and push to a public remote.

2. Paste the URL for the folder into the extension

3. Run

As always, feel free to follow along in our new public repo and please reach out. Everything we’ve discussed in this blog post is available for you to try out on your own projects.

For more on what we're doing at Docker, subscribe to our newsletter.