-
Notifications
You must be signed in to change notification settings - Fork 3
Description
Feature Description
Make the Context Sync MCP server available as a packaged MCP in the Docker MCP Toolkit, enabling one-command setup and simplified onboarding for users who want to run it locally without manual environment configuration.
Problem It Solves
Right now, anyone wanting to use the MCP server must manually clone, install dependencies, configure environment variables, and run it. This creates friction for users who only want to quickly test or integrate Context Sync as an MCP—especially those unfamiliar with Python environments or local MCP configs.
Publishing the server into the Docker MCP Toolkit would eliminate most of this setup burden.
Proposed Solution
Follow Docker’s MCP Registry contribution guidelines (from docker/mcp-registry) to package and validate the Context Sync MCP server.
Once prepared, publish it to the Docker MCP catalog with:
- A compliant
mcp.jsonmanifest - A Docker-friendly entrypoint
- Clear description / metadata
- A testable MCP server that runs cleanly inside the toolkit
After publication, users should be able to run:
docker run –rm -p 3000:3000 context-sync-mcp
or select it directly from the Docker MCP Toolkit UI.
Alternatives Considered
- Shipping prebuilt Python virtualenv/uv setup scripts (still manual).
- Asking users to configure MCP via local filesystem and manual JSON manifests.
- Hosting a cloud-hosted MCP endpoint (increases operational cost and does not help local workflows).
Use Case
A developer wants to sync memory/context across LLM sessions without juggling local environment setup.
They open the Docker MCP Toolkit, search for “Context Sync,” enable it, and immediately have a running MCP server connected to their AI tools—no cloning, no installs, no config.
Additional Context
- Repo: https://github.com/Intina47/context-sync
- Docker MCP Toolkit docs: https://docs.docker.com/ai/mcp-catalog-and-toolkit/toolkit/
- Docker MCP Registry contribution guide: https://github.com/docker/mcp-registry/blob/main/CONTRIBUTING.md
Providing a Docker-ready MCP package will make onboarding dramatically easier for developers who want to experiment or integrate this into their agent workflows.
Would You Use This?
Yes—this directly streamlines setup and lowers friction for local testing, development, and demo environments.
Priority
- Critical - Blocks my workflow
- High - Would significantly improve my experience
- Medium - Nice to have
- Low - Just an idea