Automated labeling of GitHub issues with Lisp and LLMs

2 hours ago 2

An AI-powered GitHub Issue labeler

gail will review the Issues in your GitHub-hosted project and, where appropriate, assign new labels to the Issues from a list of label options you provide. gail is currently configured to use OpenAI LLMs to accomplish this.

I wrote gail to add some semblance of order to libffi's ~200 open Issues. It exceeded expectations, and I hope you enjoy it as well!

Download all dependencies using ocicl.

Now run make. You are done.

Labeling Issues has never been easier. Here's the gail help message with all of the instructions that you'll need:

NAME: gail - GitHub Issue Auto-Labeler using AI USAGE: gail OWNER REPO OPTIONS: --help display usage information and exit --version display version and exit -l, --labels <PATH> Labels file [default: .gail-labels] -m, --model <VALUE> OpenAI model to use [default: gpt-4o-mini] -n, --dry-run Show what would be labeled without actually labeling EXAMPLES: Label issues in the libffi/libffi repository: gail libffi libffi Dry run with custom labels file: gail --labels my-labels.txt --dry-run microsoft vscode Use a different OpenAI model: gail --model gpt-4 owner repo AUTHORS: Anthony Green <[email protected]> LICENSE: MIT License

Note that you can simply commit your custom .gail-labels file to your repo and run gail in your project's root directory in order to pick up your custom labels. As an example, here's the file I created for libffi: https://github.com/libffi/libffi/blob/master/.gail-labels

gail was written by Anthony Green, and is distributed under the terms of the MIT license.

Read Entire Article