Research

Paper

AI LLM March 18, 2026

TDAD: Test-Driven Agentic Development - Reducing Code Regressions in AI Coding Agents via Graph-Based Impact Analysis

Authors

Pepe Alonso

Abstract

AI coding agents can resolve real-world software issues, yet they frequently introduce regressions, breaking tests that previously passed. Current benchmarks focus almost exclusively on resolution rate, leaving regression behavior under-studied. This paper presents TDAD (Test-Driven Agentic Development), an open-source tool and benchmark methodology that combines abstract-syntax-tree (AST) based code-test graph construction with weighted impact analysis to surface the tests most likely affected by a proposed change. Evaluated on SWE-bench Verified with two local models (Qwen3-Coder 30B on 100 instances and Qwen3.5-35B-A3B on 25 instances), TDAD's GraphRAG workflow reduced test-level regressions by 70% (6.08% to 1.82%) and improved resolution from 24% to 32% when deployed as an agent skill. A surprising finding is that TDD prompting alone increased regressions (9.94%), revealing that smaller models benefit more from contextual information (which tests to verify) than from procedural instructions (how to do TDD). An autonomous auto-improvement loop raised resolution from 12% to 60% on a 10-instance subset with 0% regression. These findings suggest that for AI agent tool design, surfacing contextual information outperforms prescribing procedural workflows. All code, data, and logs are publicly available at https://github.com/pepealonso95/TDAD.

Metadata

arXiv ID: 2603.17973
Provider: ARXIV
Primary Category: cs.SE
Published: 2026-03-18
Fetched: 2026-03-19 06:01

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.17973v1</id>\n    <title>TDAD: Test-Driven Agentic Development - Reducing Code Regressions in AI Coding Agents via Graph-Based Impact Analysis</title>\n    <updated>2026-03-18T17:38:22Z</updated>\n    <link href='https://arxiv.org/abs/2603.17973v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.17973v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>AI coding agents can resolve real-world software issues, yet they frequently introduce regressions, breaking tests that previously passed. Current benchmarks focus almost exclusively on resolution rate, leaving regression behavior under-studied. This paper presents TDAD (Test-Driven Agentic Development), an open-source tool and benchmark methodology that combines abstract-syntax-tree (AST) based code-test graph construction with weighted impact analysis to surface the tests most likely affected by a proposed change. Evaluated on SWE-bench Verified with two local models (Qwen3-Coder 30B on 100 instances and Qwen3.5-35B-A3B on 25 instances), TDAD's GraphRAG workflow reduced test-level regressions by 70% (6.08% to 1.82%) and improved resolution from 24% to 32% when deployed as an agent skill. A surprising finding is that TDD prompting alone increased regressions (9.94%), revealing that smaller models benefit more from contextual information (which tests to verify) than from procedural instructions (how to do TDD). An autonomous auto-improvement loop raised resolution from 12% to 60% on a 10-instance subset with 0% regression. These findings suggest that for AI agent tool design, surfacing contextual information outperforms prescribing procedural workflows. All code, data, and logs are publicly available at https://github.com/pepealonso95/TDAD.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SE'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-18T17:38:22Z</published>\n    <arxiv:comment>Toolpaper, 7 pages, 3 tables, 1 figure, 1 algorithm. Submitted to ACM AIWare 2026 (Data and Benchmark Track)</arxiv:comment>\n    <arxiv:primary_category term='cs.SE'/>\n    <author>\n      <name>Pepe Alonso</name>\n    </author>\n  </entry>"
}