Skill Packs bundle everything you have taught your AI into a single link your team can install in one command. Here is the setup, and the audit to run before you install someone else's.
You spend months teaching AI how you work, what your business does, and how you want things handed back. Then a teammate asks how you did it and there is no good answer. You send four links, a zip file, and a paragraph of instructions. They install two things and give up.
On August 7, 2026, skills.sh shipped Skill Packs, which fixes exactly that. You bundle your skills into one pack and it gives you a single link. The other person runs one command and has your setup.
The genuinely useful part is that it is not locked to one AI. You can use Claude, someone on your team uses Cursor, someone else uses Codex, and the same skills work for all three. Here is how to install one, how to build yours, and the checklist to run before you install anyone else's.
One
A Pack Can Pull From Anywhere
You are not limited to skills already published on the directory.
What you can bundle: community skills from skills.sh, folders and files sitting on your own laptop, a zip, and public or private GitHub repos. That means the half-finished skill in your documents folder can go in the same pack as a polished public one, and the person installing it never knows the difference.
Two
Every Pack Is Unlisted
Packs do not show up in a public directory. Each one gets its own URL and you decide who sees it.
What that means for you: you can put your real business context in a pack and share it with three people without publishing it to the internet. Packs are owned by a Vercel account, and Vercel teams are the groups you can share to, so you will sign in with Vercel to make one.
Three
One File, Every Agent
The same skill installs into whichever AI tool the person is already using.
Where it lands: Claude Code reads .claude/skills/ in your project and ~/.claude/skills/ globally. Codex reads .agents/skills/ and ~/.codex/skills/. Cursor reads .agents/skills/ and ~/.cursor/skills/. The installer knows all of them, so nobody on your team has to care where the files go.
You do not install the tool first. It runs through npx, which means the command downloads it, uses it, and gets out of the way. Open your terminal in the project folder you want the skills in, then:
Then restart your AI tool so it picks up the new files. If a skill does not seem to be firing, that restart is the reason nine times out of ten.
A skill is not a document. It is a set of instructions your AI will follow, with whatever access you have already given it. Installing a stranger's pack is closer to running their script than to reading their blog post, and packs being unlisted means there is no public rating or install count to lean on.
None of that means avoid them. It means spend ninety seconds first. The checklist below is short on purpose, so you will actually use it.
The reason to build one is not tidiness. It is that everyone on your team stops getting different results from the same request.
If you want a place to find skills worth putting in a pack, I walked through the free directory and how to spot the good ones in the skills marketplace guide.
The Version Problem Nobody Warns You About
Once people install your pack, they have a copy, not a live link. Change a skill later and their version is quietly out of date. Put a date at the top of every skill file and tell your team when you have updated the pack, or you will spend a month debugging why one person's results look different.