Backend

LaTeX Chatbot Server

A FastAPI-based backend for the LaTeX Chatbot Editor. This server handles LaTeX project management, compilation (via Tectonic), and AI-powered chat interactions.

Features

• Project Management: Initialize, read, and update LaTeX projects.
• Compilation: Compile LaTeX files to PDF using Tectonic.
• AI Chat: Context-aware chat using LangGraph agents.
• Configuration: Manage user settings and LLM credentials.

graph TD
    API[FastAPI Server] --> PM[Project Management]
    API --> C[Compilation]
    API --> Chat[AI Chat]
    API --> Config[Configuration]

    PM --> FS[(File System)]
    C --> Tectonic{Tectonic Engine}
    Chat --> LG[LangGraph Agent]
    LG --> LLM((LLM Provider))
    Chat --> FS

Loading

API Reference

Project Ops

• POST /init: Initialize a new project in a directory.
• GET /files: List files in a project directory.
• GET /files/content: internal endpoints for reading files.
• PUT /files/content: internal endpoints for updating files.

Compilation

• POST /compile: Trigger a compilation for a project.
• GET /pdf: Retrieve the compiled PDF.

Chat & AI

• POST /chat: Send a message to the AI agent.
• POST /copilotkit: Stream events for CopilotKit integration.

Configuration

• GET /config: Retrieve current configuration.
• POST /config: Update configuration (e.g., OpenAI keys).
• POST /nuke: Reset configuration to defaults.

Development

The server is built with Python 3.10+ and managed by uv.

Running Locally

# From project root
make server

Or manually:

cd editor-cli
uv run latex-chatbot-server

The server runs on http://127.0.0.1:8765 by default.