Short answer
An agent package manager stores reusable instructions for AI agents. It helps an agent find the right workflow, install it, and use it inside a project.
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.
What this means
Agents need instructions
An AI agent can write code, read files, and call tools. But it still needs clear instructions. A package can tell the agent how your team reviews code, writes tests, or handles releases.
Packages make workflows repeatable
If every project has a different hidden prompt, the agent behaves differently each time. A package gives the workflow a name and a version.
AIPM keeps it project-local
AIPM installs files into the project. That means the workflow can live near the code and can be reviewed like other project changes.
Simple steps to create and AI agent
- Write one clear agent workflow.
- Put the workflow in a skill file.
- Publish it as an AIPM package.
- Install it into each project that needs it.
- Update the package when the workflow changes.
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
Does an agent package manager run the agent?
No. It manages the files and instructions that the agent uses. Your AI tool still runs the agent.
Can one package work for many agents?
Yes. The package can include a different file for each AI tool that it supports. AIPM installs the right file for each tool.