Fix #34: Docker image

[JiwaniZakir]

Closes #34

Before

Memora could only be run by installing it directly in a Python environment. There was no containerized option, requiring users to manage Python 3.12 dependencies and environment configuration manually.

After

Memora can be run as a Docker container. The image uses python:3.12-slim as a base, installs Memora from pyproject.toml, and starts the server via the memora-server entrypoint. Transport defaults to SSE, listening on 0.0.0.0:8000, with persistent storage mounted at /data/memora.db.

Changes

• Dockerfile (new file):
  • Copies pyproject.toml and the memora/ package directory, then runs pip install --no-cache-dir . to install the package and its dependencies.
  • Sets four environment variables as defaults: MEMORA_TRANSPORT=sse, MEMORA_HOST=0.0.0.0, MEMORA_PORT=8000, MEMORA_STORAGE_URI=/data/memora.db. All can be overridden at runtime via -e.
  • Declares /data as a named volume so the SQLite database persists across container restarts.
  • Exposes port 8000 and runs CMD ["memora-server"].

Testing

Build and run locally:

docker build -t memora .
docker run -p 8000:8000 -v memora-data:/data memora

Verified the server starts and listens on port 8000. Storage persists across container restarts when using the /data volume mount.

---

This PR was created with AI assistance (Claude). The changes were reviewed by quality gates and a critic model before submission.