What is chatnest?

chatnest is an open source, placeholder-only version of the author's personal AI chat web app. The public repository intentionally leaves out the real logo, custom fonts, private prompts, environment secrets, databases, uploaded files, memory data, and logs, since those pieces cannot legally be shared. Anyone using the project needs to download a separate asset pack and swap in their own logo, icons, and fonts, using materials they own or are licensed to use. There are two ways to run it, and the README suggests picking based on what you want. A frontend-only demo folder shows the interface using fake data, with no backend or API key needed, useful for just looking at or modifying the UI. A full-stack folder contains the complete app, both frontend and backend together, since the interface and backend are closely tied together, for example when displaying the model's step by step reasoning. The full-stack version includes features like switching between AI models, streaming responses, showing tool usage and reasoning summaries, file uploads, saved conversation history, a memory interface, and profile and preference settings. The full-stack version also includes an optional local memory retrieval service. It searches a user's own local text memory files using a combination of vector search through ChromaDB and keyword search using jieba and BM25, without needing any extra API key, and the vector database itself is generated locally on the user's machine from their own memory files, such as a CLAUDE.md file, a profile file, and a memories folder. Getting the frontend demo running is as simple as starting a basic Python web server in that folder and opening it in a browser, or just double clicking the HTML file. Running the full-stack version involves setting up a Python virtual environment, installing dependencies, copying an example environment file, generating a secret key, setting a password, and then running a startup script, with a separate script available if the optional memory service should also be enabled. The project is licensed for non-commercial use only: copying, modifying, and redistributing it for non-commercial purposes is allowed, commercial use is not, and giving credit to the author is appreciated but not required.