# Inside the Scaffold: A Source-Code Taxonomy of Coding Agent Architectures

arXiv: https://arxiv.org/abs/2604.03515 · HTML: https://mlsanity.com/paper/2604.03515/

## Key facts

- Cited by 18 papers 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 28 of its 61 references that are in the map.
- First posted on arXiv 3 Apr 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, 28 of 28)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [Gorilla: Large Language Model Connected with Massive APIs](https://mlsanity.com/paper/2305.15334/) (2023)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?](https://mlsanity.com/paper/2509.16941/) (2025)
- [RepairAgent: An Autonomous, LLM-Based Agent for Program Repair](https://mlsanity.com/paper/2403.17134/) (2024)
- [The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Survey](https://mlsanity.com/paper/2404.11584/) (2024)
- [The BrowserGym Ecosystem for Web Agent Research](https://mlsanity.com/paper/2412.05467/) (2024)
- [Understanding Software Engineering Agents: A Study of Thought-Action-Result Trajectories](https://mlsanity.com/paper/2506.18824/) (2025)
- [A Survey on Large Language Models for Software Engineering](https://mlsanity.com/paper/2312.15223/) (2023)
- [A Survey of Learning-based Automated Program Repair](https://mlsanity.com/paper/2301.03270/) (2023)
- [MASAI: Modular Architecture for Software-engineering AI Agents](https://mlsanity.com/paper/2406.11638/) (2024)
- [Understanding Code Agent Behaviour: An Empirical Study of Success and Failure Trajectories](https://mlsanity.com/paper/2511.00197/) (2025)
- [What's in a GitHub Star? Understanding Repository Starring Practices in a Social Coding Platform](https://mlsanity.com/paper/1811.07643/) (2018)
- [DARS: Dynamic Action Re-Sampling to Enhance Coding Agent Performance by Adaptive Tree Traversal](https://mlsanity.com/paper/2503.14269/) (2025)
- [SWE-Effi: Re-Evaluating Software AI Agent System Effectiveness Under Resource Constraints](https://mlsanity.com/paper/2509.09853/) (2025)
- [Understanding Automated Program Repair Agents Through the Lens of Traceability: An Empirical Study](https://mlsanity.com/paper/2506.08311/) (2025)
- [Saving SWE-Bench: A Benchmark Mutation Approach for Realistic Agent Evaluation](https://mlsanity.com/paper/2510.08996/) (2025)
- [Rethinking the Value of Agent-Generated Tests for LLM-Based Software Engineering Agents](https://mlsanity.com/paper/2602.07900/) (2026)
- [Prometheus: Towards Long-Horizon Codebase Navigation for Repository-Level Problem Solving](https://mlsanity.com/paper/2507.19942/) (2025)
- [SWE-Compass: Towards Unified Evaluation of Agentic Coding Abilities for Large Language Models](https://mlsanity.com/paper/2511.05459/) (2025)
- [Harness Engineering for Agentic AI Coding Tools: An Exploratory Study](https://mlsanity.com/paper/2602.14690/) (2026)
- [Toward Architecture-Aware Evaluation Metrics for LLM Agents](https://mlsanity.com/paper/2601.19583/) (2026)

## Built on by (most cited first, 18 of 18)

- [HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?](https://mlsanity.com/paper/2609.01437/) (2026)
- [Harness Engineering: Anatomy, Architecture, and Evolution of Coding Agents -- A Source-Code Study of Eleven Systems](https://mlsanity.com/paper/2609.00006/) (2026)
- [An Empirical Study of Harness Design for Coding Agents](https://mlsanity.com/paper/2609.20804/) (2026)
- [Don't Blame the Large Language Model: How Agent Harness Evolution Shapes Coding Agent Quality](https://mlsanity.com/paper/2607.03691/) (2026)
- [LoopArena: Benchmarking Models as Runtime Controllers for Loop Engineering](https://mlsanity.com/paper/2608.28281/) (2026)
- [Writing Bug Reports for Software Repair Agents: What Information Matters Most?](https://mlsanity.com/paper/2607.09553/) (2026)
- [When Does Restricting a Coding Agent to executecode Help? A Regime × Agent-Design Ablation](https://mlsanity.com/paper/2607.10569/) (2026)
- [StarHarness: Evolving Harnesses with Stratified Search for Enterprise Environments](https://mlsanity.com/paper/2608.24804/) (2026)
- [PYTHALAB-MERA: Validation-Grounded Memory, Retrieval, and Acceptance Control for Frozen-LLM Coding Agents](https://mlsanity.com/paper/2605.08468/) (2026)
- [Rethinking Complexity Metrics for LLM-Integrated Applications: Beyond Source Code](https://mlsanity.com/paper/2607.01903/) (2026)
- [From Prompt to Harness: Coderlet from Scratch](https://mlsanity.com/paper/2608.09480/) (2026)
- [Understanding the Architecture of Coding Agents: An Exploratory Study Using a Research Prototype](https://mlsanity.com/paper/2608.10934/) (2026)
- [The Empire, Long Divided, Must Unite: Architectural Convergence in Three LLM Agent Harnesses](https://mlsanity.com/paper/2608.23953/) (2026)
- [The reach of a verification tool decides its value: A controlled study of verification surface, artifact quality, and cost in AI coding agents](https://mlsanity.com/paper/2608.28795/) (2026)
- [Hearsay: Can an Auditor Trust the Record a Deployed Agent Harness Writes?](https://mlsanity.com/paper/2609.32495/) (2026)
- [What Does a Harness Buy? Tokens, Mostly](https://mlsanity.com/paper/2610.04433/) (2026)
- [Does an Agent's History Tell You When Compaction Will Hurt? A Modest, Bounded Effect on the TRACE Paired-Replay Corpus](https://mlsanity.com/paper/2610.08722/) (2026)
- [A Decade in a Day: Synthesizing Agent-Native Interfaces from Human-Oriented Software Evolution](https://mlsanity.com/paper/2610.15765/) (2026)

## Abstract (arXiv)

LLM-based coding agents can localize bugs, generate patches, and run tests with diminishing human oversight, yet the scaffolding code that surrounds the language model (the control loop, tool definitions, state management, and context strategy) remains poorly understood. Existing surveys classify agents by abstract capabilities (tool use, planning, reflection) that cannot distinguish between architecturally distinct systems, and trajectory studies observe what agents do without examining the scaffold code that determines why. This paper presents a source-code-level architectural taxonomy derived from analysis of 13 open-source coding agent scaffolds at pinned commit hashes. Each agent is characterized across 12 dimensions organized into three layers: control architecture, tool and environment interface, and resource management. The analysis reveals that scaffold architectures resist discrete classification: control strategies range from fixed pipelines to Monte Carlo Tree Search, tool counts range from 0 to 37, and context compaction spans seven distinct strategies. Five loop primitives (ReAct, generate-test-repair, plan-execute, multi-attempt retry, tree search) function as composable building blocks that agents layer in different combinations; 11 of 13 agents compose multiple primitives rather than relying on a single control structure. Dimensions converge where external constraints dominate (tool capability categories, edit formats, execution isolation) and diverge where open design questions remain (context compaction, state management, multi-model routing). All taxonomic claims are grounded in file paths and line numbers, providing a reusable reference for researchers studying agent behavior and practitioners designing new scaffolds.