AgentForge: Execution-Grounded Multi-Agent LLM Framework for Autonomous Software Engineering

TL;DR

AgentForge introduces an execution-grounded multi-agent framework for autonomous software engineering, emphasizing sandboxed code verification to improve correctness and performance.

Contribution

This work formalizes execution-grounded verification as a core principle and demonstrates its effectiveness in a multi-agent LLM system for software development.

Findings

Achieves 40.0% resolution on SWE-BENCH Lite, surpassing single-agent baselines by 26-28 points.

Execution feedback and role decomposition independently improve system performance.

Open-source implementation available at https://github.com/raja21068/AutoCodeAI.

Abstract

Large language models generate plausible code but cannot verify correctness. Existing multi-agent systems simulate execution or leave verification optional. We introduce execution-grounded verification as a first-class principle: every code change must survive sandboxed execution before propagation. We instantiate this principle in AGENTFORGE, a multi-agent framework where Planner, Coder, Tester, Debugger, and Critic agents coordinate through shared memory and a mandatory Docker sandbox. We formalize software engineering with LLMs as an iterative decision process over repository states, where execution feedback provides a stronger supervision signal than next-token likelihood. AGENTFORGE achieves 40.0\% resolution on SWE-BENCH Lite, outperforming single-agent baselines by 26--28 points. Ablations confirm that execution feedback and role decomposition each independently drive performance. The framework is open-source at https://github.com/raja21068/AutoCodeAI.