Short answer
Teams should keep safe MCP config and setup notes in Git, keep secret values local, and package reusable MCP instructions with AIPM.
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.
- Repository or repo: a project folder whose changes are saved and tracked.
- Git: a tool that records file changes so people can review or undo them.
- Config: settings that tell a tool how to work.
- MCP: a standard way for an AI tool to connect to other tools and information.
- Environment variable: a private setting stored on your computer, often used for a password or token.
What this means
MCP config connects agents to tools
MCP lets AI coding agents use external tools and data sources. That makes setup powerful, but it also means config needs review.
Do not commit secrets
A safe package can explain which server to use and which environment variables are needed. It should not include real tokens, passwords, or private values.
Use packages for repeated setup
Several projects may need the same MCP setup. Put the safe settings and notes in an AIPM package. Each project can add its own private values.
Simple steps to create and AI agent
- Write what the MCP server does.
- List the safe config files.
- List required environment variables without values.
- Add a test step so users can confirm setup.
- Package reusable setup with AIPM.
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 mcp.json be shared in Git?
Yes, if it does not include secrets. Keep private values in local environment variables or ignored files.
Why use AIPM for MCP setup?
AIPM helps teams install the same safe setup notes and instructions across many repos without copying them by hand.