Are Coding Agents Generating Over-Mocked Tests? An Empirical Study

• URL: http://arxiv.org/abs/2602.00409v1
• Date: Fri, 30 Jan 2026 23:55:25 GMT
• Title: Are Coding Agents Generating Over-Mocked Tests? An Empirical Study
• Abstract summary: Coding agents have received significant adoption in software development recently.<n>This paper presents the first study to investigate the presence of mocks in agent-generated tests of real-world software systems.<n>Overall, we find that coding agents are more likely to modify tests and to add mocks to tests than non-coding agents.
• Score: 2.3625700564650347
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Coding agents have received significant adoption in software development recently. Unlike traditional LLM-based code completion tools, coding agents work with autonomy (e.g., invoking external tools) and leave visible traces in software repositories, such as authoring commits. Among their tasks, coding agents may autonomously generate software tests; however, the quality of these tests remains uncertain. In particular, excessive use of mocking can make tests harder to understand and maintain. This paper presents the first study to investigate the presence of mocks in agent-generated tests of real-world software systems. We analyzed over 1.2 million commits made in 2025 in 2,168 TypeScript, JavaScript, and Python repositories, including 48,563 commits by coding agents, 169,361 commits that modify tests, and 44,900 commits that add mocks to tests. Overall, we find that coding agents are more likely to modify tests and to add mocks to tests than non-coding agents. We detect that (1) 60% of the repositories with agent activity also contain agent test activity; (2) 23% of commits made by coding agents add/change test files, compared with 13% by non-agents; (3) 68% of the repositories with agent test activity also contain agent mock activity; (4) 36% of commits made by coding agents add mocks to tests, compared with 26% by non-agents; and (5) repositories created recently contain a higher proportion of test and mock commits made by agents. Finally, we conclude by discussing implications for developers and researchers. We call attention to the fact that tests with mocks may be potentially easier to generate automatically (but less effective at validating real interactions), and the need to include guidance on mocking practices in agent configuration files.

Related papers

• ExecCritic: Learn to Test, Test to Improve for Coding Agents [68.42713285082912]
  Execution feedback can guide coding agents toward correct repository repairs, but only when the tests capture the behavior requested by the issue.<n>We introduce ExecCritic, combining a test--verify--revise scaffold with a role-specific reinforcement learning recipe for training agents within it.<n>The scaffold separates test construction from source-code repair: a Test agent independently generates repository-native tests, a fail-closed harness qualifies and freezes them, and a Repair agent revises source code from their execution feedback without changing the tests.
  arXiv  Detail & Related papers  (2026-09-08T17:53:37Z)
• Test Coverage Analysis of Agentic Pull Requests [13.366199469803922]
  coding agents submit complete pull requests (PRs) with minimal human intervention.<n>As these agents become more prevalent, ensuring the code they generate is adequately tested is critical to preventing regressions.<n>We analyze 4882 agent-generated PRs from the AIDev dataset (532 Java and 4350 Python PRs) produced by five coding agents.
  arXiv  Detail & Related papers  (2026-07-20T15:26:30Z)
• SWE-chat: Coding Agent Interactions From Real Users in the Wild [70.18158706281724]
  SWE-chat is the first large-scale dataset of real coding agent sessions collected from open-source developers in the wild.<n>The dataset currently contains 6,000 sessions, comprising more than 63,000 user prompts and 355,000 agent tool calls.
  arXiv  Detail & Related papers  (2026-04-22T17:08:19Z)
• SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks [55.76734816061826]
  We introduce SlopCodeBench, a language-agnostic benchmark comprising 20 problems and 93 checkpoints.<n>We track two trajectory-level quality signals: verbosity, the fraction of redundant or duplicated code, and structural erosion.<n>No agent solves any problem end-to-end across 11 models; the highest checkpoint solve rate is 17.2%.
  arXiv  Detail & Related papers  (2026-03-25T19:26:44Z)
• Testing with AI Agents: An Empirical Study of Test Generation Frequency, Quality, and Coverage [1.6782288954198918]
  This study presents an empirical analysis of test generation by agent-based coding tools using the AIDev dataset.<n>We extracted 2,232 commits containing test-related changes and investigated three aspects: the frequency of test additions, the structural characteristics of the generated tests, and their impact on code coverage.<n>Our findings reveal that (i) AI authored 16.4% of all commits adding tests in real-world repositories, (ii) AI-generated test methods exhibit distinct structural patterns, featuring longer code and a higher density of assertions, and (iii) AI-generated tests contribute to code coverage comparable to human-written tests
  arXiv  Detail & Related papers  (2026-03-14T03:01:44Z)
• Rethinking the Value of Agent-Generated Tests for LLM-Based Software Engineering Agents [20.29427807019999]
  Large Language Model (LLM) code agents increasingly resolve repository-level issues by iteratively editing code, invoking tools, and validating candidate patches.<n>In these, agents often write tests on the fly, a paradigm adopted by many high-ranking agents on the SWE-bench leaderboard.<n>This raises the critical question: whether such tests meaningfully improve issue resolution or merely mimic human testing practices while consuming a substantial interaction budget.
  arXiv  Detail & Related papers  (2026-02-08T10:26:31Z)
• Holistic Agent Leaderboard: The Missing Infrastructure for AI Agent Evaluation [87.47155146067962]
  We provide a standardized evaluation harness that orchestrates parallel evaluations across hundreds of tasks.<n>We conduct three-dimensional analysis spanning models, scaffolds, and benchmarks.<n>Our analysis reveals surprising insights, such as higher reasoning effort reducing accuracy in the majority of runs.
  arXiv  Detail & Related papers  (2025-10-13T22:22:28Z)
• Intention-Driven Generation of Project-Specific Test Cases [45.2380093475221]
  We propose IntentionTest, which generates project-specific tests given the description of validation intention.<n>We extensively evaluate IntentionTest against state-of-the-art baselines (DA, ChatTester, and EvoSuite) on 4,146 test cases from 13 open-source projects.
  arXiv  Detail & Related papers  (2025-07-28T08:35:04Z)
• SWE-PolyBench: A multi-language benchmark for repository level evaluation of coding agents [49.73885480071402]
  We introduce SWE-PolyBench, a new benchmark for repository-level, execution-based evaluation of coding agents.<n>SWE-PolyBench contains 2110 instances from 21 repositories and includes tasks in Java (165), JavaScript (1017), TypeScript (729) and Python (199), covering bug fixes, feature additions, and code.<n>Our experiments show that current agents exhibit uneven performances across languages and struggle with complex problems while showing higher performance on simpler tasks.
  arXiv  Detail & Related papers  (2025-04-11T17:08:02Z)
• CLOVER: A Test Case Generation Benchmark with Coverage, Long-Context, and Verification [71.34070740261072]
  This paper presents a benchmark, CLOVER, to evaluate models' capabilities in generating and completing test cases.<n>The benchmark is containerized for code execution across tasks, and we will release the code, data, and construction methodologies.
  arXiv  Detail & Related papers  (2025-02-12T21:42:56Z)
• Commit0: Library Generation from Scratch [77.38414688148006]
  Commit0 is a benchmark that challenges AI agents to write libraries from scratch.<n>Agents are provided with a specification document outlining the library's API as well as a suite of interactive unit tests.<n> Commit0 also offers an interactive environment where models receive static analysis and execution feedback on the code they generate.
  arXiv  Detail & Related papers  (2024-12-02T18:11:30Z)
• Evaluating Software Development Agents: Patch Patterns, Code Quality, and Issue Complexity in Real-World GitHub Scenarios [13.949319911378826]
  This study evaluated 4,892 patches from 10 top-ranked agents on 500 real-world GitHub issues.<n>No single agent dominated, with 170 issues unresolved, indicating room for improvement.<n>Most agents maintained code reliability and security, avoiding new bugs or vulnerabilities.<n>Some agents increased code complexity, many reduced code duplication and minimized code smells.
  arXiv  Detail & Related papers  (2024-10-16T11:33:57Z)
• TestGenEval: A Real World Unit Test Generation and Test Completion Benchmark [24.14654309612826]
  TestGenEval comprises 68,647 tests from 1,210 code and test file pairs across 11 well-maintained Python repositories.<n>It covers initial tests authoring, test suite completion, and code coverage improvements.<n>We evaluate several popular models, with sizes ranging from 7B to 405B parameters.
  arXiv  Detail & Related papers  (2024-10-01T14:47:05Z)
• SWT-Bench: Testing and Validating Real-World Bug-Fixes with Code Agents [10.730852617039451]
  We investigate the capability of LLM-based Code Agents to formalize user issues into test cases.<n>We propose a novel benchmark based on popular GitHub repositories, containing real-world issues, ground-truth bug-fixes, and golden tests.<n>We find that LLMs generally perform surprisingly well at generating relevant test cases, with Code Agents designed for code repair exceeding the performance of systems designed for test generation.
  arXiv  Detail & Related papers  (2024-06-18T14:54:37Z)
• Observation-based unit test generation at Meta [52.4716552057909]
  TestGen automatically generates unit tests, carved from serialized observations of complex objects, observed during app execution. TestGen has landed 518 tests into production, which have been executed 9,617,349 times in continuous integration, finding 5,702 faults. Our evaluation reveals that, when carving its observations from 4,361 reliable end-to-end tests, TestGen was able to generate tests for at least 86% of the classes covered by end-to-end tests.
  arXiv  Detail & Related papers  (2024-02-09T00:34:39Z)
• AgentCoder: Multi-Agent-based Code Generation with Iterative Testing and Optimisation [11.155351560550853]
  This paper introduces Multi-Agent Assistant Code Generation (AgentCoder) AgentCoder is a novel solution comprising a multi-agent framework with specialized agents: the programmer agent, the test designer agent, and the test executor agent. Our experiments on 9 code generation models and 12 enhancement approaches showcase AgentCoder's superior performance over existing code generation models.
  arXiv  Detail & Related papers  (2023-12-20T13:22:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.