Digest.

Five July 2026 papers on coding agents converge on one practitioner lesson: quality, cost, safety, and honest evaluation live in the scaffolding and workflow around the model, not in the weights.

Five late-June papers on why coding-agent verification has to come from outside the model — from process-discipline and interactive-session benchmarks to self-review collapse, cheap environment-free verifiers, and the Codex adoption curve raising the stakes.

Five fresh June papers converge on one uncomfortable truth for anyone building AI coding tools: the agent you ship is a system, not a model — and this month's reliability wins all live in the harness, the guardrails, and the orchestration rather than the weights.

This week's arxiv crop is all multi-agent orchestration for software engineering — and a maturing skepticism, backed by complexity metrics and controlled experiments, that more agents rarely means better software.

Four papers from the past ten days converging on the same conclusion: agentic coding's bottleneck has moved from raw generation capability to process control, runtime architecture, and verifier design.

Five papers from the last ten days that all point at the same shift: the model is the easy part — the harness, the workflow store, and the way we measure rollouts are where coding agents now succeed or fail.

Six new arxiv papers on agentic coding — ProgramBench, Mise en Place, Proactivity, Constraint Decay, SWE Atlas, and Shepherd — with a practitioner's read on each.

Five papers from this week tackle the layer above 'does the model write code': repository-level repair, subagent specialization, compositional safety attacks, full-program synthesis, and a compiler for the SKILL.md format.