Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Humans Integrate, Agents Fix: How Agent-Authored Pull Requests Are Referenced in Practice

Islem Khemissi, Moataz Chouchen, Dong Wang, Raula Gaikovina Kula

Abstract

Although coding agents have introduced new coordination dynamics in collaborative software development, detailed interactions in practice remain underexplored, especially for the code review process. In this study, we mine agent-authored PR references from the AIDev dataset and introduce a taxonomy to characterize the intent of these references across Human-to-Agent and Agent-to-Agent interactions in the form of Pull Requests (i.e. PRs). Our analysis shows that while humans initiate most references to agent-authored PRs, a substantial portion of these interactions are AI-assisted, indicating the emergence of meta-collaborative workflows, where humans mostly use references to build new features, whereas agents make them to fix errors. We further find that referencing/referenced PRs are associated with substantially longer lifespans and review times compared to isolated PRs, suggesting higher coordination or integration effort. We then list three key takeaways as potential future research directions into how to utilize these dynamics for optimizing AI coding agents in the code review process.

Humans Integrate, Agents Fix: How Agent-Authored Pull Requests Are Referenced in Practice

Abstract

Although coding agents have introduced new coordination dynamics in collaborative software development, detailed interactions in practice remain underexplored, especially for the code review process. In this study, we mine agent-authored PR references from the AIDev dataset and introduce a taxonomy to characterize the intent of these references across Human-to-Agent and Agent-to-Agent interactions in the form of Pull Requests (i.e. PRs). Our analysis shows that while humans initiate most references to agent-authored PRs, a substantial portion of these interactions are AI-assisted, indicating the emergence of meta-collaborative workflows, where humans mostly use references to build new features, whereas agents make them to fix errors. We further find that referencing/referenced PRs are associated with substantially longer lifespans and review times compared to isolated PRs, suggesting higher coordination or integration effort. We then list three key takeaways as potential future research directions into how to utilize these dynamics for optimizing AI coding agents in the code review process.

Paper Structure

This paper contains 7 sections, 5 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Study Design
  3. Experimental Results
  4. RQ1: To what degree are agent-authored PRs linked in the code review process?
  5. RQ2: What are the impacts of referencing agent-authored PRs on the code review process?
  6. RQ3: What are the primary drivers of interaction with agent-authored PRs?
  7. Takeaways and Future Agenda

Figures (5)

  • Figure 1: Example of PR referencing. A reference from one PR creates an event on the referenced PR's timeline (PR #30300).
  • Figure 2: Methodology
  • Figure 3: Agent-authored PRs Reference Prevalence
  • Figure 4: Human-Agent References: Solo Human vs AI-Assisted Distribution
  • Figure 5: Unified taxonomy of pull-request reference types. Percentages indicate the distribution of Agent--Agent (A--A) and Human--Agent (H--A) interactions within each category.