⚠️ Alpha Release Disclaimer

LingoLeaf is currently in Alpha. It is an experimental tool under active development.

• Hardware Requirements: Running local LLMs (via Ollama/Docker) requires significant CPU/GPU and RAM. Performance will vary based on your local machine.
• Liability: By using this software, you acknowledge that it is provided "as is." The developers are not responsible for any data loss, system instability, or hardware strain caused by running local models.
• No Cloud Connectivity: While we do not send your data to external servers, you are responsible for securing your own local environment and Docker instances.

---

LingoLeaf (Alpha)

LingoLeaf is in early development. Features and documentation may change between releases.

---

1. What is LingoLeaf?

What it is: A private, local-first Chrome extension that saves highlighted French text and the page URL, then asks a local LLM (Ollama in Docker on your machine) for a quick English gloss—without using a cloud translation API for that step. Your data stays on your device unless you choose to share it elsewhere.

---

2. Prerequisites

Requirement	Purpose
Docker Desktop (or equivalent Docker engine + Compose v2)	Runs the Ollama container and stores models in a persistent volume.
Google Chrome (or another Chromium browser that supports unpacked extensions)	Installs and runs the LingoLeaf extension.

You need network access only when pulling the AI model the first time (or after wiping the Docker volume). Day-to-day use of translation does not require outbound calls to a third-party API for that step.

---

3. Quick Start

From the repository root, run these two commands in order.

1. Start the local translation engine

docker compose up -d

2. Download the translation model (once per machine/volume, or after docker compose down -v)

docker compose --profile setup run --rm model-setup

The first command starts Ollama and exposes the API at http://127.0.0.1:11434. The second pulls the configured model (qwen2.5:3b) into the named volume so it survives container restarts.

Equivalent alternative: if you prefer a shell helper, ./setup_model.sh performs the same pull after checking that the Ollama service is running.

Sanity check (optional):

curl -s http://127.0.0.1:11434/api/tags

You should see your model listed. If the list is empty, run step 2 again.

---

4. Chrome installation

1. Open Chrome and go to chrome://extensions.
2. Turn on Developer mode (toggle in the top-right).
3. Click Load unpacked.
4. Select the extension folder inside this repository (the folder that contains manifest.json).

After loading, pin the extension if you like (Extensions menu → pin LingoLeaf). Open the toolbar popup to view saved items and use actions such as Fill missing when the local engine was offline earlier.

To pick up code changes after editing files, return to chrome://extensions and click Reload on LingoLeaf.

---

Additional notes

• Stop the engine: docker compose down (model files remain in the ollama_models volume unless you remove volumes with docker compose down -v).
• Configuration: The Compose file sets OLLAMA_HOST and OLLAMA_ORIGINS so the extension can reach Ollama from a chrome-extension:// origin; see docker-compose.yml for details.