Skip to content

Step 1: Setup environment

Install dependencies and choose your starting point.

Prerequisites check

python3 --version  # Should be 3.8 or higher
git --version      # Any recent version

Choose your path

Path A: Use the sample project (Recommended for first-time users)

  • No GitHub token or local repo required for steps 1–2
  • Uses a built-in Python CLI project called TaskFlow
  • Run with --sample flag

Path B: Use your own repository

  • Works with any local directory or GitHub repo
  • Requires an AI provider API key
  • Results are immediately useful

Install dependencies

From the 03-readme-generator/ directory:

pip install --upgrade pip
pip install anthropic pyyaml jinja2
# or for OpenAI:
pip install openai pyyaml jinja2
# for GitHub repos, also add:
pip install PyGithub

Configure AI access

The script reads from the shared config.yaml at the root of the repository. If you completed either of the other projects, your config is already set up.

If not, copy the example:

# From the repo root
cp config.example.yaml config.yaml

Open config.yaml and set:

ai_provider: "anthropic"      # or "openai"
ai_api_key: "sk-ant-..."      # your API key
model: "claude-sonnet-4-6"    # or your preferred model

Path A users

Skip AI configuration for now — --sample mode runs without API calls. You will need it for Step 3 when you generate actual content.

Verify your setup

cd 03-readme-generator
python readme_generator.py --sample --analyze

Expected output:

📄 README Generator
==================================================

🔍 Analyzing sample project...
✓ Project: taskflow
✓ Language: Python
✓ License: MIT
✓ Dependencies: click, pyyaml, rich
✓ Entry points: main.py

📋 Full repo context:
...

If you see this output, you are ready to continue.

Next step: Analyze your repo →