Configuration Smells in AGENTS.md Files: Common Mistakes in Configuring Coding Agents
On the map: LLM agents and tool use → Software engineering coding agents
Abstract (arXiv)
Coding agents are increasingly used to automate software engineering tasks. To guide their behavior, these agents commonly rely on configuration files, typically named AGENTS. md or CLAUDE. md, which provide instructions about architecture, workflows, coding conventions, and testing practices. Despite their growing importance, little is known about common problems affecting the definition and maintenance of these files. In this paper, we present the first catalog of smells for coding-agent configuration files. To identify such smells, we first conducted a grey literature review and a repository mining analysis. As a result, we identified six configuration smells and proposed automated heuristics to detect them. To evaluate the prevalence of the proposed smells, we analyzed 100 popular open-source repositories containing either an AGENTS. md or a CLAUDE. md file. Our results show that configuration smells are widespread. Lint Leakage was the most common smell, affecting 62% of the files, followed by Context Bloat (42%) and Skill Leakage (35%). We further show that several smells frequently co-occur, particularly Context Bloat, Skill Leakage, and Conflicting Instructions.
Key facts. Cited by 1 paper in the map (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 37 references that are in the map. First posted on arXiv 14 Jun 2026.
Builds on
21 of 37 references.
Closest ancestors · 3
Same direction, newest first
- 2025Decoding the Configuration of AI Coding Agents: Insights from Claude Code Projects12 Nov
- 2025Using Copilot Agent Mode to Automate Library Migration: A Quantitative Assessment30 Oct
- 2025On the Use of Agentic Coding Manifests: An Empirical Study of Claude Code18 Sep
Other references · 18
- 2021Evaluating Large Language Models Trained on Codecited by 8,320
- 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,373
- 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 1,439
- 2023Large Language Models for Software Engineering: A Systematic Literature Reviewcited by 399
- 2023Using an LLM to Help With Code Understandingcited by 228
- 2024Automated Unit Test Improvement using Large Language Models at Metacited by 75
- 2023Using Large Language Models to Generate JUnit Tests: An Empirical Studycited by 53
- 2023LLaMA-Reviewer: Advancing Code Review Automation with Large Language Models through Parameter-Efficient Fine-Tuningcited by 47
- 2017Guidelines for including grey literature and conducting multivocal literature reviews in software engineeringcited by 41
- 2022Using Transfer Learning for Code-Related Taskscited by 18
- 2025Migrating Code At Scale With LLMs At Googlecited by 14
- 2024PromptSet: A Programmer's Prompting Datasetcited by 9
6 more
- 2022Bug Analysis in Jupyter Notebook Projects: An Empirical Studycited by 5
- 2022Fixing Dockerfile Smells: An Empirical Studycited by 3
- 2022On the Prevalence, Impact, and Evolution of SQL Code Smells in Data-Intensive Systemscited by 2
- 2024On the Quality of AI-Generated Source Code Comments: A Comprehensive Evaluationcited by 2
- 2022Code Smells in Elixir: Early Results from a Grey Literature Reviewcited by 1
- 2026Cache-Related Smells in GitLab CI/CD: Comprehensive Catalog, Automated Detection, and Empirical Evidencecited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2026SmellBench: Evaluating LLM Agents on Architectural Code Smell Repair
- 2025Agent READMEs: An Empirical Study of Context Files for Agentic Coding
- 2026Harness Engineering for Agentic AI Coding Tools: An Exploratory Study
- 2026A Deterministic Control Plane for LLM Coding Agents
- 2026Scanning the Harness: A Repository Study of Configuration Exposures in AI Coding Agents
- 2026On the Impact of AGENTS.md Files on the Efficiency of AI Coding Agents
- 2026Trust but Verify? Uncovering the Security Debt of Autonomous Coding Agents
- 2025Agentic LMs: Hunting Down Test Smells