Skip to main content
SAICA-KG

Papers

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

Rajesh Kumar, Waqar Ali, Junaid Ahmed, Najmah Ali, Shaban Usman · 2026 · arXiv

Source →arXiv:2604.131200 citations
multi-agentexecution-groundedsandboxingcoding-agentcorrectionverification

TL;DR

Introduces execution-grounded verification as a first-class principle: every code change must survive sandboxed execution before propagation. AGENTFORGE instantiates this with Planner/Coder/Tester/Debugger/Critic agents sharing memory and a mandatory Docker sandbox, achieving 40.0% on SWE-Bench Lite — 26-28 points above single-agent baselines.

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.

Editorial notes

Must-cite. Canonical example of "sandbox as first-class verification" multi-agent architecture. Grounds SAICA-KG's correction ControlParadigm tools and the environment LocusOfControl; pairs with Yan 2025 (fault- tolerant sandboxing) for the sandbox Tool cluster.

View raw YAML on GitHub →