6 papers
ProjAgent: Procedural Similarity Retrieval for Repository-Level Code Generation
QiHong Chen, Aaron Imani, Iftekhar Ahmed
Repository-level code generation requires implementing target functions while accounting for complex cross-file dependencies and project-specific conventions. Existing retrieval me…
Agents with Feelings? Personality and Emotion in Multi-Agent Software Teams
Yunyan Ding, Thomas Zimmermann, Iftekhar Ahmed
Multi-agent LLM systems for Software Engineering (SE) typically differentiate agents through roles and workflows, but little is known about how agents' behavioral profiles affect t…
From Anatomy to Smells: An Empirical Study of SKILL.md in Agent Skills
David Boram Hong, Aaron Imani, Iftekhar Ahmed
Agent Skills provide on-demand domain knowledge to LLM agents without requiring model retraining. Each Agent Skill is defined by a mandatory SKILLmd file containing metadata and…
Prompting in the Wild: An Empirical Study of Prompt Evolution in Software Repositories
Mahan Tafreshipour, Aaron Imani, Eric Huang +3
The adoption of Large Language Models (LLMs) is reshaping software development as developers integrate these LLMs into their applications. In such applications, prompts serve as th…
Context Conquers Parameters: Outperforming Proprietary LLM in Commit Message Generation
Aaron Imani, Iftekhar Ahmed, Mohammad Moshirpour
Commit messages provide descriptions of the modifications made in a commit using natural language, making them crucial for software maintenance and evolution. Recent developments i…
Investigating the Impact of Code Comment Inconsistency on Bug Introducing
Shiva Radmanesh, Aaron Imani, Iftekhar Ahmed +1
Code comments are essential for clarifying code functionality, improving readability, and facilitating collaboration among developers. Despite their importance, comments often beco…