ProseAiC

A self-hosted document editor with integrated AI assistance. Built for people who work with documents not just code.

ProseAiC is model-agnostic and BYOK (bring your own key). It works with all major cloud AI providers, and is especially designed for homelab use with open-weight models running on your own hardware via Ollama so your documents and API keys don't have to leave your network.

---

Features

• Document editor — create, organise, and edit documents in a clean writing environment
• AI chat panel — talk to the AI assistant to get help with your writing
• Tool-using AI — the assistant can directly create, edit, move, and restructure document content through tool calls, not just make suggestions
• Model-agnostic — supports Ollama, Anthropic, OpenAI, Gemini, and Mistral
• Self-hosted — runs entirely on your own machine; no accounts, no cloud sync
• BYOK — API keys are entered in the app and stored in your browser's local storage; they are never sent to the ProseAiC server

---

Supported AI platforms

Platform	Type	Notes
Ollama	Local / Cloud	Recommended for homelab and offline use
Anthropic	Cloud	Requires API key
OpenAI	Cloud	Requires API key
Gemini	Cloud	Requires API key
Mistral	Cloud	Requires API key

To enable cloud providers, enter your API key and endpoint URL in the chat settings. For self-hosted environments proxy endpoints can be configured in the server's .env file so that requests are forwarded to the correct destination through the server.

---

Prerequisites

• Node.js 20 or later
• One or more of:
  • A running Ollama instance (local or on your network)
  • An API key from a supported cloud provider

---

Getting started

After cloning the repo, set up your environment:

npm install
cp .env.example .env

Edit .env to point at your Ollama instance (or leave the default if Ollama is running locally on port 11434).

npm start

Open http://localhost:3001 in your browser.

Open the Settings panel to configure your AI platforms

Enter the endpoint URL and API key for each provider you want to use.

Settings are saved to your browser's local storage and never sent to the server.

For Ollama, the default endpoint is https://ollama.com (their hosted platform). To use a local or homelab instance, enter its address directly in Settings (e.g. http://localhost:11434) — or set the endpoint to /ollama to route requests through the server proxy.

---

Server configuration

Server behaviour is configured via environment variables in .env. See .env.example for all options.

Variable	Default	Description
OLLAMA_HOST	https://ollama.com	Where the server proxy forwards Ollama requests (used when endpoint is set to /ollama in Settings)
OLLAMA_TIMEOUT_MS	—	Request timeout for Ollama (ms). Useful for slow hardware or large models
STORE_DIR	./documents	Directory where documents are stored on disk

---

Development

npm run watch        # build, watch for changes, and start the server
npm test             # unit tests
npm run test:smoke   # end-to-end smoke tests

See CONTRIBUTING.md for full contribution guidelines.

---

License

MIT