GET_STARTED.md

🚀 DeCodifier — Local Setup (ZIP Download Edition)

Welcome! This guide assumes you downloaded this repo as a .zip file.

1️⃣ Create a Python environment cd decodifier-main python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate

2️⃣ Install dependencies pip install --upgrade pip pip install -e . pip install openai # required for the demo client

3️⃣ Start the engine uvicorn engine.app.main:app --reload

Your local API is now running at:

http://127.0.0.1:8000

🔑 Required Keys (for Demos) export OPENAI_API_KEY="your-key-here"

⚠️ Usage Charges Warning Running demos may call external APIs. You are responsible for any usage fees from providers. DeCodifier performs no billing or metering.

🧪 Run the OpenAI Todo Demo python clients/openai_demo/decodifier_openai_demo.py

Expected behavior:

The model inspects the project

Generates a Todo feature

Wires it into the app

🔍 Test the Todo API

With the engine running:

curl http://127.0.0.1:8000/api/todos curl -X POST http://127.0.0.1:8000/api/todos -H "Content-Type: application/json" -d '{"title":"First task"}' curl -X PUT http://127.0.0.1:8000/api/todos/1/done

🛡 Troubleshooting Issue Fix KeyError: 'path' Update to latest / ensure defensive tool adapter applied ModuleNotFoundError: openai pip install openai Address in use Kill existing uvicorn or use port --port 8001 Windows path issues Use WSL or GitBash recommended 📁 Data Storage

DeCodifier stores state here:

~/.decodifier

Move or override with:

export DECODIFIER_DATA_DIR=/your/custom/dir

🎉 Next Steps

Connect to other LLMs (Claude, Groq, Gemini)

Try patch-based editing: tools → decodifier_patch_file

Add custom patterns in decodifier_patterns/

DeCodifier turns LLMs into real coding assistants — not autocomplete. Build something cool. And if you ship a product on top, tell us. We'll cheer you on. 🚀