Installation¶
repository-manager is a standard Python package and a prebuilt container image.
Pick the path that matches how you want to run it.
Requirements¶
- Python 3.11+.
- A reachable Git executable on
PATH(the package shells out togitfor bulk operations). - A workspace directory containing the Git repositories you intend to manage — set
via
REPOSITORY_MANAGER_WORKSPACE(see Deployment).
From PyPI (recommended)¶
Optional extras¶
The base install is intentionally minimal. Install the extra for what you need:
| Extra | Install | Pulls in |
|---|---|---|
mcp |
pip install "repository-manager[mcp]" |
FastMCP MCP-server runtime (agent-utilities[mcp]) |
agent |
pip install "repository-manager[agent]" |
Pydantic-AI agent + Logfire tracing, pre-commit, bump2version |
test |
pip install "repository-manager[test]" |
pytest, pytest-xdist, pytest-asyncio, pytest-cov, pytest-timeout |
all |
pip install "repository-manager[all]" |
Everything above |
# Typical: run the MCP server and the release/maintenance harness
pip install "repository-manager[all]"
From source¶
git clone https://github.com/Knuckles-Team/repository-manager.git
cd repository-manager
pip install -e ".[all]" # editable install with every extra
With uv:
Prebuilt Docker image¶
A multi-stage, slim image is published on every release (installs
repository-manager[all]):
docker pull knucklessg1/repository-manager:latest
docker run --rm -i \
-e REPOSITORY_MANAGER_WORKSPACE=/workspace \
-v /home/apps/workspace:/workspace \
knucklessg1/repository-manager:latest # stdio transport (default)
For an HTTP server with a published port, the agent server, and Docker Compose, see Deployment.
Verify the install¶
repository-manager --version
repository-manager-mcp --help
python -c "import repository_manager; print('repository-manager ready')"
Next steps¶
- Deployment — run it as a long-lived MCP / agent server behind Caddy + DNS.
- Usage — call the tools, the
Gitclient, and the CLI. - Configuration — every environment variable.