The Software Librarian: Python Package Insights for Copilot
Software packages form the backbone of software systems, significantly influencing their functionality, efficiency, and long-term maintainability. As developers increasingly turn to Large Language Models (LLMs) to streamline software development tasks, the ability of these models to accurately recommend suitable packages becomes critical. However, LLMs lack the ability to provide real-time information about package details such as license, dependencies, or even their existence. This can lead to the integration of outdated, incompatible, or legally restrictive packages, which could compromise the software's quality and legal standing. In this paper, we introduce the Software Librarian, a tool that provides real-time information about Python packages recommended as part of the generated code by GitHub Copilot, including license details, deprecation status, and package health. Our tool ensures that the recommended packages are not only valid but are also suitable for integration. To support future research, we have made the Software Librarian available on the Visual Studio Marketplace<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup><sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup>https://marketplace.visualstudio.com/items?itemName=jaslatendresse. software-librarian and released the code online.<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup><sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>https://github.com/jaslatendresse/software-librarian-prod A demonstration can be viewed at https://youtu.be/hnPr0rvL8lk.
Authors
- Emad Shihab (ORCID: https://orcid.org/0000-0003-1285-9878)
- SayedHassan Khatoonabadi
- Jasmine Latendresse (ORCID: https://orcid.org/0000-0002-0481-8331)
- Nicholas Day
Institutions
- Concordia University (CA)
- University of Sousse (TN)
Publication Details
- Published
- 2025-04-27
- DOI
- https://doi.org/10.1109/icse-companion66252.2025.00014
- Primary Topic
- Scientific Computing and Data Management
- Type
- article
- Field-Weighted Citation Impact
- 0.00