How do you guys trust AI enough to ship without checking what it built?
I'm running 5 to 8 agents in parallel, but I still need to check the final output before pushing to prod.
I know there are workflows and tools to automate this, but I can't seem to let go yet
Anyone else…
Commit
0f66d8e
I make sure tests are written and have 100% coverage and I often (but not always... depends) run
/code-review (both the Anthropic one and the one from Matt Pocock to make sure it matches all my .ai/rules)This is a great thing, I am using Larapilot at the moment, and it has review processes, testing, and everything, but still, even though it has autopilot and could do everything on its own, I still do not feel comfortable giving up control. Even though control basically means asking what it did and saying "okay".
I feel like this attitude of mine is a bit masochistic
commit 0f66d8e
Don't know what that is. I'd rather build my own workflows than use something opinionated by someone else. Try start fresh without this tool and just ask the LLM to "check tests pass 100%" coverage or [if you're using Claude] run the
code-review skill.
Build plugins/skills as…