Context Engineering for Multi-Agent LLM Code Assistants Using Elicit, NotebookLM, ChatGPT, and Claude Code
- ID: cc44c3c6
- 原文链接: https://arxiv.org/abs/2508.08322
- PDF: https://arxiv.org/pdf/2508.08322v1
- 作者: Muhammad Haseeb
- 日期: 2025-08-09
- 更新: 2025-08-09
- 分类: coding
- 来源类型: paper
- 标签: context-engineering, multi-agent, claude-code, rag, code-generation
- 质量评分: 4/5
- 抓取时间: 2026-09-19T04:21:30Z
- arXiv 备注: 15 pages, 5 figures, research paper on multi-agent LLM systems for code generation
中文导读
一套多组件 context engineering 工作流：GPT-5 做 Intent Translator 澄清需求，Elicit 语义检索注入领域知识，NotebookLM 做文档综合，Claude Code 多 agent 系统负责生成与验证在大型 Next.js 代码库上，多 agent 系统能以极少人工干预完成复杂特性的规划编辑测试，单发成功率和项目上下文遵循度都高于单 agent 基线与 CodePlanMASAIHyperAgent 对比后，作者把收益归因于定向上下文注入与 agent 角色分解两个手段
为什么值得关注
GPT-5 澄清意图 + Elicit 检索 + NotebookLM 综合 + Claude Code 干活：上下文工程四件套
关键信息
- 论文标题：Context Engineering for Multi-Agent LLM Code Assistants Using Elicit, NotebookLM, ChatGPT, and Claude Code
- 作者：Muhammad Haseeb
- arXiv：https://arxiv.org/abs/2508.08322
- 发布时间：2025-08-09
- arXiv 分类：cs.SE, cs.AI
- 关联标签：context-engineering, multi-agent, claude-code, rag, code-generation
English Abstract
Large Language Models (LLMs) have shown promise in automating code generation and software engineering tasks, yet they often struggle with complex, multi-file projects due to context limitations and knowledge gaps. We propose a novel context engineering workflow that combines multiple AI components: an Intent Translator (GPT-5) for clarifying user requirements, an Elicit-powered semantic literature retrieval for injecting domain knowledge, NotebookLM-based document synthesis for contextual understanding, and a Claude Code multi-agent system for code generation and validation. Our integrated approach leverages intent clarification, retrieval-augmented generation, and specialized sub-agents orchestrated via Claude's agent framework. We demonstrate that this method significantly improves the accuracy and reliability of code assistants in real-world repositories, yielding higher single-shot success rates and better adherence to project context than baseline single-agent approaches. Qualitative results on a large Next.js codebase show the multi-agent system effectively plans, edits, and tests complex features with minimal human intervention. We compare our system with recent frameworks like CodePlan, MASAI, and HyperAgent, highlighting how targeted context injection and agent role decomposition lead to state-of-the-art performance. Finally, we discuss the implications for deploying LLM-based coding assistants in production, along with lessons learned on context management and future research directions.
English Summary
A context engineering workflow that combines multiple AI components: an Intent Translator (GPT-5) for clarifying requirements, Elicit-powered semantic literature retrieval for injecting domain knowledge, NotebookLM-based document synthesis, and a Claude Code multi-agent system for code generation and validation. On a large Next.js codebase the multi-agent system plans, edits, and tests complex features with minimal human intervention, yielding higher single-shot success rates and better project-context adherence than baseline single-agent approaches. Compared against CodePlan, MASAI, and HyperAgent, targeted context injection and agent role decomposition drive the gains; 15 pages (cs.SE, cs.AI).
Obsidian Notes
- 内容由
opencli arxiv paper拉取 arXiv 元数据与摘要生成。 - 中文导读与价值判断均锚定在条目已有摘要、论文摘要、作者、日期与分类信息上；未补充论文摘要之外的实验细节。