The company brain for AI agents. Read context. Write artifacts. Ledger every outcome.
US
Joined March 2026
- Tweets3
- Following2
- Followers1
- Likes0
If you want to do this without using Github... our playbook feature does this out-of-the-box.
It has a few features onto of this too:
- Create bundles of skills for different teams/agents
- Reference shared documents
and it's completely backward-compatible with git.
Send this to ANYONE on your team using AI agents with Claude/Codex skills:
If I were you, I'd put your BEST AI skills in a GitHub repo, turn that repo into a PLUGIN, and have your team INSTALL it in Claude/Codex with auto update on.
Why?
1. Everyone gets the same AI SOPs instead of 10 versions floating around Slack
2. When one person improves a skill, the whole team gets the better version.
3. New hires can start with your best workflows instead of a blank AI setup (this is a BIG deal).
4. If someone breaks a skill, you can roll it back with version control.
5. Personal skills can stay personal, while team skills become shared company infrastructure (and an asset!).
6. Your best AI processes stay with the company when someone leaves.
7. You can chain skills together for bigger workflows, like titles -> thumbnails -> descriptions -> YouTube publish.
8. You can track which skills are actually being used and delete the ones collecting dust.
9. You get less slop because the agent has real instructions, examples, taste, and process.
10. Your team moves from single-player AI to multiplayer AI.
(thanks to @aiwithremy for coming onto @startupideaspod)
Watch full breakdown here (clearly explained): youtu.be/xHsftiyT9pQ?si=pcwq…
I don't know why I didn't do this before.
The more I think about it, the more obvious it feels: your AI workflows should be version controlled company assets.
It kinda feels like the difference between “we use AI” and “we actually operate with AI".
Enjoy.
11
Patina retweeted
The other day, our marketing agent produced a monthly content plan for customers we were no longer trying to reach.
It was full of ideas for local businesses. In July, we’d stopped treating local businesses as our ideal customer, so seeing them throughout the plan made no sense.
I checked the agent's context. And yeah, both versions of the ICP were still there.
I knew which version we'd adopted. But the agent didn’t. It saw two documents that both looked valid.
Human review caught this mistake because it was obvious. But the more workflows you hand to agents, the less you can rely on review to catch every outdated decision.
What was missing was the decision between the documents: the July ICP had replaced the April version.
So I’ve built a way to capture that decision in Patina. When a strategic document changes, Patina finds the conflicting references and works out what the new version replaces. I approve the change then Patina retires the old version and updates those references.
It turns out that coordinating a team of agents has the same challenges as coordinating people. In this case: making sure everyone knows what changed and what is still true.