Configuration Guide¶

This guide explains how to configure the Academic Document Generator using JSON configuration files.

📋 Overview¶

The tool supports JSON-based configuration for three use cases:

Colloquium protocols (thesis defense)
Project work & WASP1 grading letters
Peer review comments

🎯 Quick Start¶

1. Choose a Template¶

# List available templates
academic-doc-generator --list-templates

2. Copy and Customize¶

# Copy template to your thesis folder
cp config_templates/config_colloquium_campus.json /path/to/thesis/

After copying, customize the config.json file. At a minimum, you must set the correct PDF filename (pdf -> filename).

3. Run the Tool¶

Option A: Use main.py You can run the tool via main.py in the project root. To do this, edit the folder variable in main.py to point to the directory containing your PDF and the JSON config.

python main.py

Option B: Use CLI directly

academic-doc-generator --config /path/to/thesis/config_colloquium_campus.json

📝 Configuration Structure¶

Common Fields¶

`output` (optional)¶

{
  "output": {
    "folder": null,              // null = same as PDF folder
    "compile_pdf": true,         // Compile LaTeX to PDF
    "signature_file": "signature.png", // Path to signature image
    "create_feedback_mail": true // Generate feedback email (for projects)
  }
}

folder: Output directory (null = PDF's folder)
compile_pdf: Whether to compile .tex to PDF
signature_file: Path to signature image (defaults to signature.png or auto-detected in data/)
create_feedback_mail: Whether to generate student feedback (mainly used in project task)

🎓 Colloquium Configuration¶

{
  "task": "colloquium",
  "colloquium": {
    "date": "20.01.2026",
    "time": "14:00",
    "location_type": "campus",
    "room": "3.217"
  }
}

Metadata Overrides (Optional): You can add a metadata section to manually set fields if automatic extraction fails: - course_of_study: e.g., "Informatik", "Medieninformatik" - author: Student name

📂 Project Configuration¶

{
  "task": "project",
  "project": {
    "mark": "1.3"
  },
  "output": {
    "create_feedback_mail": true
  }
}

📄 Review Configuration¶

{
  "task": "review",
  "pdf": {
    "filename": "paper.pdf"
  }
}

🔑 API Keys Configuration¶

Create secrets.env in project root:

OPENAI_API_KEY=sk-xxxxxxxx
GROQ_API_KEY=gsk-xxxxxxxx
GEMINI_API_KEY=AIzaSyxxxxxxxx

💡 Advanced Options¶

Rate Limiting for Free Tiers¶

{
  "llm": {
    "api_choice": "groq",
    "groq_free": true
  }
}

Fill Form Only Mode¶

{
  "output": {
    "fill_form_only": true
  }
}

📝 Configuration Templates¶

Pre-built JSON configurations for common workflows in the config_templates/ directory:

config_colloquium_campus.json - Thesis colloquium on campus
config_colloquium_company.json - Thesis colloquium at a company
config_colloquium_online.json - Online thesis colloquium (Zoom)
config_project_template.json - Project work grading
config_wasp1_template.json - WASP1 project grading
config_review_template.json - Peer review comments

⚙️ Global Configuration (config.yaml)¶

In addition to project-specific JSON files, there is a global config.yaml in the project root for general settings:

# Global folder for web metadata (*.md profiles)
web_metadata_folder: "/path/to/your/website/data/projects/"

# Default first examiner name (overrides PDF extraction)
first_examiner: "Prof. Dr. Firstname Lastname"

web_metadata_folder: If set, generated Jekyll-compatible project profiles are automatically copied to this folder.
first_examiner: A name set here will be used as the first examiner for all generated documents.