{ "schema_version": 1, "builder_plan": { "scenario": "Your coding assistant makes changes that still need substantial repair before merging.", "change": "Try the technique on held-out issues from a repository with reliable tests.", "baseline": "The same coding agent, repository snapshot, tools, and token budget.", "steps": [ "Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.", "Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.", "Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.", "Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test." ], "measure": "Correct patches, regression failures, human repair time, and total cost per accepted change.", "decision": "Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation." }, "builder_review": null, "arxiv_id": "2609.26847", "title": "Who Finishes the Job? A Study of Follow-Up Fixes and Commit Authorship on AI Coding Agent Pull Requests", "summary": "This is an empirical measurement study rather than an adoptable method; it replaces intuition about whether merged agent PRs are finished work with verified fix-attribution data across five coding agents. It costs nothing to apply directly, but operationally it implies that adopting AI coding agents requires retained post-merge review and follow-up capacity rather than treating merges as complete. The finding can fail to transfer: results derive from popular open-source repositories (500+ stars) with public PR workflows, and fix rates may differ in private codebases with different review rigor, test coverage, or agent usage patterns.", "rationale": "The reader running coding agents in production should keep human review and post-merge monitoring in place, since the paper shows agent merges need fixing more often than human merges even though agents mostly perform those fixes themselves.", "technique": "Post-merge fix analysis of agent-authored PRs", "family": "agentes_de_codigo", "recommendation": "observar", "published": "2026-09-22", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.26847", "independent_implementations": 0, "total_implementations": 1, "stars": 0, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "This is an empirical measurement study rather than an adoptable method; it replaces intuition about whether merged agent PRs are finished work with verified fix-attribution data across five coding agents. It costs nothing to apply directly, but operationally it implies that adopting AI coding agents requires retained post-merge review and follow-up capacity rather than treating merges as complete. The finding can fail to transfer: results derive from popular open-source repositories (500+ stars) with public PR workflows, and fix rates may differ in private codebases with different review rigor, test coverage, or agent usage patterns.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.26847", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "Merged AI-agent PRs attract verified follow-up fixes at 1.62x the odds of contemporaneous human PRs in the same repositories; 69.6% of those fixes are authored by the same agent and 76.4% of fix PRs are agent-authored throughout.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.26847", "source_page": null, "source_excerpt": "" } ], "exposure_map": [ { "dimension": "quality", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compute", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "latency", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "operations", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compatibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "security", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "data_and_training", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "reproducibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" } ], "risks": [], "minimum_test": [], "open_questions": [], "independent_tests": [], "lab_reproductions": [], "report_available": false, "validation_tier": "unknown", "evidence_tier": "unknown", "infrastructure_basis": "unknown", "equations": [], "equations_status": "not_fetched", "core_kind": "", "equations_fetched_at": "", "equations_source_url": "", "equations_source_sha256": "", "exposures_reviewed_at": "", "exposures_source_sha256": "" }