PaperDebugger: A Plugin-Based Multi-Agent System for In-Editor Academic Writing, Review, and Editing

2025-12-02

AI Summary

Existing academic writing assistants operate as external tools, lacking deep integration with LaTeX editors (e.g., Overleaf) regarding document state, structural semantics, and version history—resulting in limited context awareness and poor synchronization. Method: We propose the first embedded multi-agent academic writing assistance system, built atop the Model Context Protocol (MCP) toolchain to enable plug-in integration of literature retrieval, citation localization, document scoring, and revision pipelines. A Chrome extension ensures bidirectional editor synchronization; Kubernetes-native orchestration enables fine-grained agent scheduling; and diff-driven updates, secure state management, and localized structured editing ensure robust operation. Contribution/Results: The system delivers parallel agent execution and context-aware automation—writing and reviewing—within a minimally intrusive UI. Early user engagement is high, empirically validating both the technical feasibility and practical utility of editor-native intelligent assistance for scholarly authoring.

Technology Category

• Multiagent Systems: Teamwork
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Humans and AI: Planning and Decision Support for Human-Machine Teams

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Large language models are increasingly embedded into academic writing workflows, yet existing assistants remain external to the editor, preventing deep interaction with document state, structure, and revision history. This separation makes it impossible to support agentic, context-aware operations directly within LaTeX editors such as Overleaf. We present PaperDebugger, an in-editor, multi-agent, and plugin-based academic writing assistant that brings LLM-driven reasoning directly into the writing environment. Enabling such in-editor interaction is technically non-trivial: it requires reliable bidirectional synchronization with the editor, fine-grained version control and patching, secure state management, multi-agent scheduling, and extensible communication with external tools. PaperDebugger addresses these challenges through a Chrome-approved extension, a Kubernetes-native orchestration layer, and a Model Context Protocol (MCP) toolchain that integrates literature search, reference lookup, document scoring, and revision pipelines. Our demo showcases a fully integrated workflow, including localized edits, structured reviews, parallel agent execution, and diff-based updates, encapsulated within a minimal-intrusion user interface (UI). Early aggregated analytics demonstrate active user engagement and validate the practicality of an editor-native, agentic writing assistant. More details about this demo and video could be found at https://github.com/PaperDebugger/PaperDebugger.

Problem

• Integrates LLM-driven reasoning directly into LaTeX editors like Overleaf
• Enables context-aware operations within the editor via multi-agent system
• Provides in-editor academic writing, review, and editing with minimal UI intrusion

Innovation

• In-editor plugin-based multi-agent system for academic writing
• Chrome extension with Kubernetes orchestration and MCP toolchain
• Bidirectional sync, version control, and secure state management