AI coding agents are now submitting pull requests (PRs) to software projects, acting not just as assistants but as autonomous contributors. As these agentic contributions are rapidly increasing across real repositories, little is known about how they behave in practice and why many of them fail to be merged. In this paper, we conduct a large-scale study of 33k agent-authored PRs made by five coding agents across GitHub. (RQ1) We first quantitatively characterize merged and not-merged PRs along four broad dimensions: 1) merge outcomes across task types, 2) code changes, 3) CI build results, and 4) review dynamics. We observe that tasks related to documentation, CI, and build update achieve the highest merge success, whereas performance and bug-fix tasks perform the worst. Not-merged PRs tend to involve larger code changes, touch more files, and often do not pass the project’s CI/CD pipeline validation. (RQ2) To further investigate why some agentic PRs are not merged, we qualitatively analyze 600 PRs to derive a hierarchical taxonomy of rejection patterns. This analysis complements the quantitative findings in RQ1 by uncovering rejection reasons not captured by quantitative metrics, including lack of meaningful reviewer engagement, duplicate PRs, unwanted feature implementations, and agent misalignment. Together, our findings highlight key socio-technical and human-AI collaboration factors that are critical to improving the success of future agentic workflows.

Abstract Conversational large-language models (LLMs), such as ChatGPT, are extensively used for issue resolution tasks, particularly for generating ideas to implement new features or resolve bugs. However, not all developer-LLM conversations are useful for effective issue resolution and it is still unknown what makes some of these conversations not helpful. In this paper, we analyze 686 developer-ChatGPT conversations shared within GitHub issue threads to identify characteristics that make these conversations effective for issue resolution. First, we empirically analyze the conversations and their corresponding issue threads to distinguish helpful from unhelpful conversations. We begin by categorizing the types of tasks developers seek help with (e.g., code generation , bug identification and fixing , test generation ), to better understand the scenarios in which ChatGPT is most effective. Next, we examine a wide range of conversational, project, and issue-related metrics to uncover statistically significant factors associated with helpful conversations. Finally, we identify common deficiencies in unhelpful ChatGPT responses to highlight areas that could inform the design of more effective developer-facing tools. We found that only 62% of the ChatGPT conversations were helpful for successful issue resolution. Among different tasks related to issue resolution, ChatGPT was most helpful in assisting with code generation, and tool/library/API recommendations, but struggled with generating code explanations. Our conversational metrics reveal that helpful conversations are shorter, more readable, and exhibit higher semantic and linguistic alignment. Our project metrics reveal that larger, more popular projects and experienced developers benefit more from ChatGPT’s assistance. Our issue metrics indicate that ChatGPT is more effective on simpler issues characterized by limited developer activity and faster resolution times. These typically involve well-scoped technical problems such as compilation errors and tool feature requests. In contrast, it performs less effectively on complex issues that demand deep project-specific understanding, such as system-level code debugging and refactoring. The most common deficiencies in unhelpful ChatGPT responses include incorrect information and lack of comprehensiveness. Our findings have wide implications including guiding developers on effective interaction strategies for issue resolution, informing the development of tools or frameworks to support optimal prompt design, and providing insights on fine-tuning LLMs for issue resolution tasks.

Large language models (LLMs) have become essential in software development, especially for issue resolution. However, despite their widespread use, significant challenges persist in the quality of LLM responses to issue resolution queries. LLM interactions often yield incorrect, incomplete, or ambiguous information, largely due to knowledge gaps in prompt design, which can lead to unproductive exchanges and reduced developer productivity.In this paper, we analyze 433 developer-ChatGPT conversations within GitHub issue threads to examine the impact of prompt knowledge gaps and conversation styles on issue resolution. We identify four main knowledge gaps in developer prompts: Missing Context, Missing Specifications, Multiple Context, and Unclear Instructions. Assuming that conversations within closed issues contributed to successful resolutions while those in open issues did not, we find that ineffective conversations contain knowledge gaps in $44.6 \%$ of prompts, compared to only $12.6 \%$ in effective ones. Additionally, we observe seven distinct conversational styles, with Directive Prompting, Chain of Thought, and Responsive Feedback being the most prevalent. We find that knowledge gaps are present in all styles of conversations, with Missing Context being the most repeated challenge developers face in issue-resolution conversations.Based on our analysis, we identify key textual and code-related heuristics—Specificity, Contextual Richness, and Clarity—that are associated with successful issue closure and help assess prompt quality. These heuristics lay the foundation for an automated tool that can dynamically flag unclear prompts and suggest structured improvements. To test feasibility, we developed a lightweight browser extension prototype for detecting prompt gaps, that can be easily adapted to other tools within developer workflows.