Context Engineering for Multi-Agent LLM Code Assistants: A GPT-5, Elicit, NotebookLM, and Claude Code Workflow

Forum topic · ✨步子哥 · 2025-12-11

Summary

A poster from Muhammad Haseeb (Virginia Tech, August 2025) presents a context engineering workflow for improving LLM-based code assistants on complex, multi-file repositories. The system combines four components: a GPT-5-based Intent Translator that converts user requirements into structured task specifications, Elicit-powered semantic retrieval over academic papers and documentation, NotebookLM-based synthesis of retrieved materials, and a Claude Code multi-agent orchestrator with specialized planner, coder, tester, and reviewer sub-agents backed by a vector database for code context. On a 180K-line test repository, the approach reportedly achieved a 3.5x higher single-shot success rate, 42% better context adherence, and a 90% reduction in human intervention, with a 68.2% success rate versus CodePlan (45.3%), HyperAgent (52.7%), and MASAI (28.3%). Future work includes CI/CD integration, large-scale context management, and cost optimization.

Context Engineering for Multi-Agent LLM Code Assistants

*Using Elicit, NotebookLM, ChatGPT, and Claude Code*

Author: Muhammad Haseeb • Virginia Tech • August 2025

Abstract

Large Language Models (LLMs) have shown promise in automating code generation, yet they struggle with complex, multi-file projects due to context limitations. This work proposes a context engineering workflow combining multiple AI components: an Intent Translator (GPT-5), Elicit-powered semantic literature retrieval, NotebookLM-based document synthesis, and a Claude Code multi-agent system. The approach leverages intent clarification, retrieval-augmented generation, and specialized sub-agents to improve the accuracy and reliability of code assistants in real-world repositories.

Problem Statement

LLMs struggle with complex, multi-file projects due to context limitations
Single-agent approaches often produce incomplete or incorrect solutions
Knowledge gaps when confronted with unfamiliar APIs or frameworks
Static context files cannot capture all relevant details for every possible task

Proposed Solution

A novel context engineering workflow combining multiple AI components
Intent clarification, retrieval-augmented generation, and specialized sub-agents
Orchestrated via Claude's agent framework with role decomposition
Targeted context injection for better adherence to project context

Key Components

1. Intent Translator (GPT-5) — Clarifies user requirements and translates them into structured task specifications for the multi-agent system. 2. Elicit Semantic Retrieval — Performs semantic search over academic papers, documentation, and Q&A resources to inject domain knowledge. 3. NotebookLM Synthesis — Creates concise summaries of retrieved materials and answers follow-up questions for detailed understanding. 4. Claude Code Multi-Agent — Orchestrates specialized sub-agents (planner, coder, tester, reviewer) with a vector database for code context.

Results & Performance

| Metric | Value | |---|---| | Higher single-shot success rate | 3.5× | | Better context adherence | 42% | | Lines of code in test repository | 180K | | Reduction in human intervention | 90% |

Comparison with Other Frameworks

| Framework | Approach | Success Rate | Key Advantage | |---|---|---|---| | This System | Context engineering + multi-agent | 68.2% | Targeted context injection | | CodePlan | Multi-step planning | 45.3% | Structured approach | | MASAI | Modular architecture | 28.3% | Specialized sub-agents | | HyperAgent | Team of agents | 52.7% | Human-like workflow |

Implications & Future Work

Production-ready deployment with CI/CD integration
Context management strategies for large-scale projects
Cost optimization for multi-agent systems
Extension to other software engineering domains