🌙 WITT — Stateless LangGraph Chatbot

This is a minimal project using LangGraph to implement a stateless chatbot. The bot uses a single tool get_current_time() when asked for the current time and responds to all other inputs using an LLM. It supports both OpenAI and local models via Ollama.

---

🚧 What it does

• Stateless LangGraph flow — no memory or database.
• One tool:

  def get_current_time() -> dict:
      """Return the current UTC time in ISO-8601 format.
      Example → {"utc": "2025‑05‑21T06:42:00Z"}"""

• Two nodes:
  • time_node — called if user asks about time.
  • llm_node — replies to all other messages.
• Compatible with OpenAI and Ollama (local LLM).

---

🛠 How to Run

git clone https://github.com/NeewMeta88/WITT
cd witt

python -m venv .venv
source .venv/bin/activate     # or .venv\Scripts\activate on Windows

pip install -r requirements.txt
langgraph dev

This starts a local dev server with the graph in memory.

---

☁️ Using OpenAI

1. Create your own OpenAI API key:

   • Visit https://platform.openai.com/account/api-keys
   • Sign in or create an account
   • Click “Create new secret key”
   • Copy and save the key somewhere safe

2. Update the existing .env file in the project root directory:

   Replace the placeholder with your actual API key:

   OPENAI_API_KEY=sk-...your_actual_key_here...
   
   

---

✅ Spec Compliance

• ✅ Stateless LangGraph chat flow
• ✅ Uses exactly one tool (get_current_time)
• ✅ No memory or database
• ✅ Works with langgraph dev
• ✅ Minimal repo: one .py file, requirements.txt, and README.md

---

🧪 Example

Input:

What time is it?

Output:

{"utc": "2025-05-29T15:04:00Z"}

Any other input is processed normally by the LLM.