Replication Package: Agent-Authored Pull Request Rejections (MSR 2026)

Karla Gonzalez's replication package for the MSR 2026 paper 'When Bots Get the Boot' contains data and scripts to reproduce empirical results on agent-authored pull request rejections. The 133.8 MB package includes a curated dataset of pull requests labeled by outcome, derived feature sets, and analysis scripts. It was last updated on May 6, 2026.

Use Cases

• Analyze structural signals associated with PR rejection based on code quality and complexity features.
• Investigate process history signals for PR outcomes based on collaboration dynamics data.
• Study socio-technical factors in AI coder integration based on labeled pull request outcomes.
• Reproduce statistical tests and figures from the associated research paper using the provided notebooks.

Strengths

• Includes all data, scripts, and documentation required to reproduce the paper's empirical results.
• Dataset is 133.8 MB in size and contains multiple derived feature sets.
• Released under the permissive CC-BY-4.0 license.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Curated from the AIDev dataset, as mentioned in the description.

Freshness

Last updated 2026-05-06 20:57:44; freshness should be verified.