cs.AINov 18, 2026

Metrics-in-the-Loop: Complexity Feedback Reshapes Agent-Written Code

Authors: Nicolas Chapados, Jean-François Gagné, Nouha Dziri

Organizations: Nera Software · Mila – Quebec Artificial Intelligence Institute · Cohere Labs · ELLIS Institute Tübingen

Abstract

Coding agents now write whole codebases in long autonomous runs, guided mainly by whether their tests pass. Yet passing tests alone does not reveal accumulating complexity or gaps in test coverage, which then persist as agents extend their own code. We introduce metrics-in-the-loop, an inference-time intervention that gives agents in-flight feedback: cheap, automatically computed code metrics injected into the agent's context during the trajectory. After each code-changing agent turn, we insert short feedback messages naming changed functions whose cyclomatic complexity exceeds a limit (complexity feedback) or changed lines that are skipped by tests (coverage feedback), giving the agent the opportunity to refactor code or strengthen tests before continuing. We run extensive experiments and ablations comparing several interventions across three long-horizon coding benchmarks and four strong models. We find that complexity feedback significantly reduces erosion, a measure of complexity concentration of code, without affecting held-out correctness of the code. Broadly, our experiments investigate a practical form of agent "re-railing", where timely, actionable measurements redirect agents before weaknesses become the foundation they keep building on. As agents scale to codebases with millions of lines, such interventions become critical to sustaining effective development and long-term code maintainability.

Figures & tables

Appendix figures & tables11 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Benchmarking Coding Agents under Evolving Requirements

    Oct 26, 2026Jia Feng, Alyssa Wu, Cuiyun Gao +3Coding Agent BenchmarksCode Generation

  2. SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks

    Mar 25, 2026Gabriel Orlanski, Devjeet Roy, Alexander Yun +7AI Coding AgentsCoding Agent Benchmarks

  3. LoopsBench: From Harness Engineering to Loop Engineering in Coding Agent Evaluation

    Jul 31, 2026Han Li, Zhemin Fang, Rili Feng +8AI Coding AgentsSoftware Engineering Benchmarks