MuseWeb is an experimental, prompt-driven web server that streams HTML straight from plain-text prompts using a large-language model (LLM). Originally built "just for fun," it currently serves as a proof-of-concept for what prompt-driven websites could become once local LLMs are fast and inexpensive. Even in this early state, it showcases the endless possibilities of minimal, fully self-hosted publishing.
Version 1.1.0 introduces a complete modular architecture for improved maintainability and extensibility.
- Prompt → Page – Point MuseWeb to a folder of
.txtprompts; each prompt becomes a routable page. - Live Reloading for Prompts – Edit your prompt files and see changes instantly without restarting the server.
- Streaming Responses – HTML is streamed token-by-token for instant first paint.
- Backend Agnostic – Works with either:
- Ollama (default, runs everything locally), or
- Any OpenAI-compatible API (e.g. OpenAI, Together.ai, Groq, etc.).
- Single Binary – Go-powered, ~7 MB static binary, no external runtime.
- Zero JS by Default – Only the streamed HTML from the model is served; you can add your own assets in
public/. - Modular Architecture – Clean separation of concerns with dedicated packages for configuration, server, models, and utilities.
- Configurable via
config.yaml– Port, model, backend, prompt directory, and OpenAI credentials. - Environment Variable Support – Falls back to
OPENAI_API_KEYif not specified in config or flags. - Thinking Tag Support – Special handling for models that support "thinking" tags (e.g., DeepSeek and r1-1776).
- Detailed Logging – Comprehensive logging of prompt file loading and request handling for easy debugging.
# 1. Clone and build
$ git clone https://github.com/kekePower/museweb.git
$ cd museweb
$ GO111MODULE=on go build .
# 2. (Optional) pull an LLM with Ollama
$ ollama pull llama3
# 3. Run with defaults (localhost:8080)
$ ./musewebOpen http://localhost:8080 in your browser. Navigation links are generated from the prompt filenames.
Copy config.example.yaml to config.yaml and tweak as needed:
server:
address: "127.0.0.1" # Interface to bind to (e.g., 127.0.0.1 or 0.0.0.0)
port: "8080" # Port for HTTP server
prompts_dir: "./prompts" # Folder containing *.txt prompt files
debug: false # Enable debug logging
model:
backend: "ollama" # "ollama" or "openai"
name: "llama3" # Model name to use
enable_thinking: false # Enable thinking tag for supported models
openai:
api_key: "" # Required when backend = "openai"
api_base: "https://api.openai.com/v1" # Change for other providersConfiguration can be overridden with CLI flags:
# Example with command-line flags
./museweb -port 9000 -model mistral -backend ollama -debug
# View all available options
./museweb -hFor OpenAI API keys, MuseWeb will check these sources in order:
- Command-line flag (
-api-key) - Configuration file (
config.yaml) - Environment variable (
OPENAI_API_KEY)
- Place text files in the prompts directory –
home.txt,about.txt, etc. - The filename (without extension) becomes the route:
about.txt → /about. system_prompt.txtis the only file that must exist. Define your site's core rules, output protocols, and structural requirements here.layout.txtis a special file that gets appended to the system prompt for all pages. Use it to define global layout, styling, and interactive elements that should be consistent across all pages.layout.min.txtis an optional alternative tolayout.txtthat produces minified HTML output, saving tokens and reducing response size. The server will use this file instead oflayout.txtif it exists.- All prompt files are loaded from disk on every request, so you can edit them and see changes without restarting the server.
- The prompt files included in this repo are examples only—update or replace them to suit your own site.
- HTML, Markdown, or plain prose inside the prompt will be passed verbatim to the model – sanitize accordingly before publishing.
- For best results, keep design instructions in
layout.txtand focus content instructions in individual page prompts.
As of version 1.1.0, MuseWeb has been fully modularized with a clean separation of concerns:
/
├── main.go # Application entry point and orchestration
├── config.yaml # Configuration file
├── public/ # Static files served directly
├── prompts/ # Prompt text files
└── pkg/ # Go packages
├── config/ # Configuration loading and validation
├── models/ # AI model backends (Ollama and OpenAI)
├── server/ # HTTP server and request handling
└── utils/ # Utility functions for output processing
- Configuration: The
configpackage handles loading settings from YAML with sensible defaults. - Model Abstraction: The
modelspackage provides a common interface for different AI backends. - HTTP Server: The
serverpackage manages HTTP requests, static file serving, and prompt processing. - Utilities: The
utilspackage contains functions for sanitizing and processing model outputs.
- Fork the repo and create a feature branch.
- Run
go vet ./... && go test ./...before opening a PR. - Follow Conventional Commits for commit messages.
Bug reports and feature ideas are very welcome! 🙏
MuseWeb is distributed under the terms of the Apache License, Version 2.0. See the LICENSE file for full details.