😸 Give your agent eyes

gm legends, happy Friday.

Codex Plugins is turning good setups into something you can actually install instead of rebuild, Agentation is for the very real pain of knowing exactly what is wrong on the screen and still having to explain it to the agent like it has never seen a button before, and Claude Code auto-fix is built for that annoying PR phase where the work is done but the cleanup is not.

Codex Plugins package reusable Codex workflows into installable bundles. Teams can bundle skills, app integrations, and MCP server setups together, then reuse the same setup across projects instead of rebuilding it every time.

🔥 Our Take: This makes sense, but it also feels like the beginning of a very familiar mess. The useful part is obvious: one person figures out the good Codex setup, everyone else can just install it. The less fun part is we are probably about five minutes away from people comparing plugin stacks like it is 2017 again.

Imed ran a seven-day experiment where Claude wrote all the code with no human review, and it went about as calmly as you would expect. It shipped some working features fast, then started breaking unrelated parts of the product, driving up API costs, dropping old data, hardcoding stuff that should have been configurable, and taking the site down with a dependency update. The post is basically a very clean reminder that code generation is not the same thing as judgment. 

Agentation turns UI feedback into something coding agents can actually use. Click any element on a page, add a note, and it gives you structured output with selectors, file paths, component hierarchy, styles, and your feedback, so Claude Code, Codex, Cursor, or whatever else is not left guessing which blue button you meant.

🔥 Our Take: This fixes a very stupid but very real problem. You can see exactly what is wrong on the screen, but the agent still has to go wandering through the codebase trying to figure out where that thing lives. Being able to point at the problem directly is way more useful than writing another paragraph about it.

Claude Code auto-fix watches your pull requests in the cloud and responds to CI failures or review comments on its own. It pushes a fix when the change is obvious, asks when it is not, and keeps running even after you close the laptop.

🔥 Our Take: Pull requests have this stupid phase where the real work is done and you are just playing whack-a-mole with red checks and reviewer nits. I would not hand it the political comments or anything architectural, but for cleanup duty this is pretty dead-on.