© 2026 Chiranjeev (@chiranr19) — All Rights Reserved. This project is source-available for viewing only; it is not open source. No copying, reuse, modification, deployment, or redistribution of any part of it (or its underlying ideas) without prior written permission — see LICENSE and SIGNATURE. Prospective employers and collaborators are welcome to read the code. · authorship sigil 6YHJ·BKOP·VJCG·NOAX

ChippyInn Booking Bot

A conversational hotel-room booking assistant. It parses natural-language requests ("a room in Chennai for 2 guests, Aug 20 to 23, under ₹1500"), keeps per-session memory so it doesn't re-ask what you've already answered, and returns matching rooms.

How it works

flowchart TD
    U([User message]) --> C[Flask /chat]
    C --> L[LLM slot extraction<br/>Llama-3 via OpenRouter]
    C --> H[Heuristics<br/>city · guests · date range]
    L --> S[Session memory<br/>fills missing slots]
    H --> S
    S -->|missing info| Q[Ask a follow-up question]
    S -->|all slots filled| SR{Search backend}
    SR -->|Meilisearch up| M[Meilisearch]
    SR -->|otherwise| J[In-memory rooms.json]
    M --> R([Room results])
    J --> R
    style R fill:#FF5CAA,color:#fff

Loading

• LLM slot-filling extracts location, dates, guests, and budget from free text.
• Heuristic fallbacks for city, guest count, and date-range parsing keep it useful even when the model is unsure.
• Resilient search — uses Meilisearch when it's running, and otherwise falls back to a built-in in-memory filter over rooms.json, so it runs from a clean clone with zero extra infrastructure.

Example

You:  I need a room in Chennai for 2 people
Bot:  What dates will you be staying (check-in and check-out)?
You:  Aug 20 to 23, budget 1500
Bot:  • Suite #1 - ChippyInn — ₹1388 | Chennai | sleeps 2

Run

python -m venv venv && source venv/bin/activate   # Windows: venv\Scripts\activate
pip install -r requirements.txt

export OPENROUTER_API_KEY=<your key>              # Windows: set OPENROUTER_API_KEY=<your key>
python server.py

Then open http://localhost:5000 and start chatting. Get an OpenRouter key at https://openrouter.ai/keys. The key is read from the OPENROUTER_API_KEY environment variable — it is never hard-coded or committed.

Optional: Meilisearch backend

In-memory search is enabled by default. To use Meilisearch instead, start it first:

docker run -p 7700:7700 getmeili/meilisearch

On startup the server prints which backend it selected.

What's inside

File	Purpose
server.py	Flask backend: LLM extraction, session memory, room search (Meili or in-memory)
rooms.json	Sample room inventory
static/	Minimal chat front-end (index.html, chat-widget.html, script.js)

License

Proprietary — All Rights Reserved. Source-available for viewing only; not open source. No use, copy, or reuse without written permission. See LICENSE.