Version control for everything: why AI agents need git for non-code tasks
Agentic coding tools have taken off, but non-programming use cases lag because they lack version control. The author argues that without git-like guardrails, LLMs can't safely edit docs, calendars, or issues. He proposes a proxy layer for staging changes or moving everything into git, noting that Jane Street's code-review-in-comments workflow makes LLM involvement trivial. The post concludes that better version control benefits humans too, and points to local-first software and Irmin as relevant resources.
It’s not just agent-style LLMs that would benefit from this integration, _I_ would be more productive if all of my tools had branches, version history, and atomic changes.
- cfjgvjh
I really really want to do version control for everything, but most of my data is binary so it really doesn't agree with git; I tried using LFS as well but it didn't work for my specific workflow.
Hopefully something less text oriented comes along in the future.
Lore looked interesting for this purpose.
- Klaster_1
> Design docs from Google Docs to checked-in markdown?
This worked pretty well for the team I'm in. Design docs in Google Docs were really hard to keep in sync with decisions and not properly agent accessible. Initially, we were concerned lack of comments would be an issue, but this didn't really came true - a Slack channel does a good enough job. With this flow, the design author makes a decision and review results, agent propagates it to individual areas, and from those to tasks (those are in MD too), and then copies descriptions to Jira. No more "we changed a thing but missed one place that depends on it", or at least not as bad as before.
- hoistway
The revert safety net is what makes agentic coding tolerable. Without version control, every AI mistake becomes permanent, and that alone kills the workflow.