# Cross-Provider Review as a Runtime Contract for Coding Agents: A Controlled Pilot and Fault-Injection Study

arXiv: https://arxiv.org/abs/2610.10961 · HTML: https://mlsanity.com/paper/2610.10961/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 21 of its 23 references that are in the map.
- First posted on arXiv 7 Oct 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 21 of 21)

- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance](https://mlsanity.com/paper/2305.05176/) (2023)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [RouteLLM: Learning to Route LLMs with Preference Data](https://mlsanity.com/paper/2406.18665/) (2024)
- [Agentless: Demystifying LLM-based Software Engineering Agents](https://mlsanity.com/paper/2407.01489/) (2024)
- [Humans or LLMs as the Judge? A Study on Judgement Biases](https://mlsanity.com/paper/2402.10669/) (2024)
- [Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2406.07791/) (2024)
- [AutoMix: Automatically Mixing Language Models](https://mlsanity.com/paper/2310.12963/) (2023)
- [When To Solve, When To Verify: Compute-Optimal Problem Solving and Generative Verification for LLM Reasoning](https://mlsanity.com/paper/2504.01005/) (2025)
- [AI Harness Engineering: A Runtime Substrate for Foundation-Model Software Agents](https://mlsanity.com/paper/2605.13357/) (2026)
- [Rethinking Code Review Workflows with LLM Assistance: An Empirical Study](https://mlsanity.com/paper/2505.16339/) (2025)
- [Cross-Model LLM Code Review: Should you use Claude to review Codex or vice versa?](https://mlsanity.com/paper/2607.21656/) (2026)
- [From Prompts to Contracts: Harness Engineering for Auditable Enterprise LLM Agents](https://mlsanity.com/paper/2607.08028/) (2026)
- [When to Screen, When to Bypass: LLM-Judges in Resource-Scarce AI-Human Workflow](https://mlsanity.com/paper/2603.13870/) (2026)
- [Think Again or Think Longer? Selective Verification for Budget-Aware Reasoning](https://mlsanity.com/paper/2606.19808/) (2026)
- [Evidence-Calibrated Runtime Reconstruction for Agent Skills Across Heterogeneous Coding Agents](https://mlsanity.com/paper/2608.08793/) (2026)
- [Silent Failures in Agentic Security Evaluation: A Validated Harness for Tool-Call Mediation Under Indirect Prompt Injection](https://mlsanity.com/paper/2609.32691/) (2026)
- [Groundability, Not Scale Alone: When Weak Reviewers Can Audit Strong Coding Agents](https://mlsanity.com/paper/2610.01023/) (2026)
- [When Does a Second Model Help? Cross-Model Review in LLM Verification](https://mlsanity.com/paper/2610.01471/) (2026)

## Abstract (arXiv)

Coding agents increasingly share a workstation while drawing on separate providers and subscription allowances. A second agent can inspect a completed answer, but the call spends another pool and may provide no substantive finding. We describe an advisory cross-provider review contract: distinct resource pools, bounded execution, restricted reviewer capabilities, complete input delivery, usable semantic output, explicit failure states and durable per-attempt evidence. In a controlled, agent-authored pilot of 20 paired development turns, eight had a material reviewer finding (95% exact interval 19.1-63.9%). A boundary-condition scan across both reviewer backends reproduced a previously discovered false success on partial input: four truncation levels passed historically and failed after repair. The scan also found and repaired cancellation during process reaping. In real CLI probes, Claude had no writing tools; Codex attempted writes in five of five read-only trials, each write tool failed, and no disposable repository changed. These tests cover specified paths and versions, not field reliability. A preregistered shadow study of metadata-only review allocation accrued 25 formal observations before an exact-runtime regression found a third defect: a reviewer exiting nonzero with a well-formed verdict was counted as complete. Exit status was not recorded per attempt, so exposure cannot be resolved retrospectively. The 25 formal and two pending records remain an audit cohort; the measurement-valid cohort restarted at zero and collection has begun. No gate result is reported.