# Dockerfiles and Containerization Configurations in codebase-memory-mcp

> Discover Dockerfiles and containerization configurations in the codebase-memory-mcp repository. GetCI testing and Glama MCP sandbox integration support.

- Repository: [Martin Vogel/codebase-memory-mcp](https://github.com/DeusData/codebase-memory-mcp)
- Tags: how-to-guide
- Published: 2026-07-14

---

**The `codebase-memory-mcp` repository contains two Dockerfiles and one docker-compose manifest that support CI testing and Glama MCP sandbox integration.**

The `codebase-memory-mcp` repository provides optional **Dockerfiles and containerization configurations** to ensure reproducible testing environments and sandboxed execution workflows. While the tool operates natively without containerization, these assets mirror the GitHub Actions CI setup and enable Glama AI score-badge validation.

## Test Infrastructure Dockerfile

The `test-infrastructure/Dockerfile` defines an image that replicates the Ubuntu CI environment using **Ubuntu 24.04** as the base. According to the repository source code, this container includes GCC, ASan/UBSan sanitizers, SQLite‑3, and Zlib—matching the dependencies required for the project's test suite.

This Dockerfile serves primarily to run the repository’s test suite in a containerized environment that exactly reproduces the GitHub Actions configuration. Developers use this to debug CI failures locally or validate changes against the canonical build environment.

To build the test infrastructure image:

```bash
docker build -t cbm-test ./test-infrastructure

```

To run the test container with the repository source mounted:

```bash
docker run --rm -v "$(pwd)":/src cbm-test

```

## Glama Sandbox Dockerfile

Located at `pkg/glama/Dockerfile`, this configuration builds a lightweight sandbox image specifically for Glama AI’s score-badge checks. The implementation fetches the statically-linked `codebase-memory-mcp` binary from the latest GitHub release and sets it as the container entrypoint.

This **Dockerfile and containerization configuration** is not required for normal operation of the tool. It exists solely to support Glama’s MCP (Model Context Protocol) sandbox requirements, allowing the AI platform to execute the binary in an isolated environment during automated scoring.

Build the Glama sandbox image:

```bash
docker build -t glama-cbm ./pkg/glama

```

Launch the sandbox with the help flag to verify the binary:

```bash
docker run --rm glama-cbm --help

```

## Docker Compose Configuration

The repository includes a [`test-infrastructure/docker-compose.yml`](https://github.com/DeusData/codebase-memory-mcp/blob/main/test-infrastructure/docker-compose.yml) file that provides a compose definition for launching the test container. Currently, this configuration references only the test-infrastructure image, but the structure allows for easy extension with auxiliary services such as databases or caching layers if future testing requirements demand them.

Start the test environment using Docker Compose:

```bash
docker compose -f test-infrastructure/docker-compose.yml up --build

```

## Building and Running Containerized Workflows

When working with these **Dockerfiles and containerization configurations**, each serves a distinct purpose in the development lifecycle:

- **CI Replication**: Use `test-infrastructure/Dockerfile` to validate changes against the exact Ubuntu 24.04 environment used in GitHub Actions.
- **Glama Integration**: Build `pkg/glama/Dockerfile` only when preparing releases for Glama AI’s sandbox validation.
- **Orchestration**: Leverage [`docker-compose.yml`](https://github.com/DeusData/codebase-memory-mcp/blob/main/docker-compose.yml) for simplified multi-service testing scenarios.

All three assets are optional; the `codebase-memory-mcp` binary compiles and executes directly on host systems without containerization requirements.

## Summary

- The repository contains two **Dockerfiles** (`test-infrastructure/Dockerfile` and `pkg/glama/Dockerfile`) and one **docker-compose.yml** file.
- The test infrastructure image mirrors the GitHub Actions CI environment with Ubuntu 24.04, GCC, ASan/UBSan, SQLite‑3, and Zlib.
- The Glama sandbox Dockerfile creates an isolated execution environment specifically for AI platform score-badge validation.
- Docker assets are optional; the tool runs natively without containerization.

## Frequently Asked Questions

### Do I need Docker to run codebase-memory-mcp?

No. The `codebase-memory-mcp` repository runs natively on Linux, macOS, and other supported platforms without Docker. The **Dockerfiles and containerization configurations** exist solely as optional aids for CI consistency and Glama AI integration, not for core functionality.

### What is the difference between the two Dockerfiles?

The `test-infrastructure/Dockerfile` creates a development and testing environment that mirrors the GitHub Actions CI setup, including all build dependencies and sanitizers. The `pkg/glama/Dockerfile` produces a minimal runtime image that downloads and executes the pre-compiled binary specifically for Glama AI’s automated scoring sandbox.

### How can I reproduce the CI environment locally?

Build and run the test infrastructure container by executing `docker build -t cbm-test ./test-infrastructure` followed by `docker run --rm -v "$(pwd)":/src cbm-test`. This mounts your local source code into a container that exactly matches the Ubuntu 24.04 CI environment used in automated testing.

### Is Kubernetes or other orchestration supported?

The repository does not include Kubernetes manifests or additional orchestration configurations. Only the [`test-infrastructure/docker-compose.yml`](https://github.com/DeusData/codebase-memory-mcp/blob/main/test-infrastructure/docker-compose.yml) compose file is provided, and it currently supports only single-container test execution.