Short answer
Package the public MCP setup notes, usage rules, and examples. Keep private tokens and secret values outside the package.
Words used in this guide
- AIPM: a tool that installs and updates reusable AI instructions.
- Package: a named group of files that people can install and update together.
- Workflow: a set of steps used to finish a task.
- Config: settings that tell a tool how to work.
- MCP: a standard way for an AI tool to connect to other tools and information.
What this means
Separate public setup from secrets
A package can explain which MCP server to use and how the assistant should use it. It should not include real secret values.
Explain when to use the server
Tell the assistant what the MCP server is for. Also say when not to use it, especially if it can access private systems.
Give users a safe checklist
A good MCP package gives users a short checklist. It says what to install, which settings to add, and how to test the connection.
Simple steps to create and AI agent
- Write what the MCP server does.
- List config files that are safe to share.
- List private values without including the values.
- Add a simple test command or check.
- Publish the safe setup notes as a package.
Where to go next
If you want to try this in a real project, start with the install guide, then read how to use AIPM. If you want to share your own workflow, read the publishing guide.
FAQ
Can I publish MCP tokens in a package?
No. Publish instructions, not secret values. Users should add their own tokens locally.
Why package MCP setup at all?
It keeps the setup consistent, so every project does not need to rebuild the same notes from scratch.