This is a LLM chat client that stores the conversations as files/folders in a git-friendly format.

Other features:

• Allows switching between different LLM providers, including locally hosted
• Features MCP support

Running the Application in Development Mode

1. Create a local environment file from the example:

cp src/chat/.env.example src/chat/.env

2. Update the values in src/chat/.env as required for your environment.

3. Start the development server:

npm run dev

4. Open http://localhost:3000

---

Running Unit Tests

Execute:

npm run test

---

Running End-to-End (E2E) Tests

You can run E2E tests using one of the following approaches.

---

Option 1 — Without Containers

This approach requires running the mock LLM API and the application server manually.

1. Start the mock LLM API (keep it running):

npm run start:mock-openai-api

2. Start the application server (keep it running):

npm run start:e2e

3. Run the E2E tests:

npm run test:e2e

---

Option 2 — Using a Container

If you are developing inside a devcontainer, ensure that the Docker-in-Docker feature is enabled.

With this approach, the mock LLM API runs as a private service alongside the application server, so you do not need to start it manually.

1. Start the containerized environment:

npm run start:e2e:container

2. Run the E2E tests:

npm run test:e2e

---

Option 3 — Running E2E Tests Against the Dev Server

E2E tests use an in-memory store, so they can run against the development server without interfering with your local development session.

Important difference in LLM configuration:

• When using start:e2e or start:e2e:container, the application uses a mock LLM API.
• When using the dev server, the application uses the LLM configuration defined in your .env file. This may cause some tests to fail depending on your configuration.

1. Start the development server:

npm run dev

2. Run the E2E tests:

npm run test:e2e

---

Using the App to Develop Itself

The ./notes directory contains prompts used during the research and development of the application.

To start the app configured to store prompts in the ./notes directory:

cd deploy
docker compose up --build

The application will be available at:

http://localhost:4000