You can write a better prompt. Your agent can make a better guess.
You are looking at a page thinking, "Move this down, use the same spacing as that card, but only on mobile." You know exactly what you mean because you're looking at it. Your coding agent is looking at a repository. Your coding agent cannot read your mind So you translate what you see into a prompt. Which element, which page, which breakpoint, this instance or every instance. Add a screenshot and…
While reviewing a webpage, you recognize the need to adjust its layout specifically for mobile view. Your coding assistant, however, lacks the ability to intuitively understand your intentions. Therefore, you must translate your observations into a prompt detailing the element, page, breakpoint, or scope you wish to modify. Including a screenshot aids the agent in comprehending the visual changes required, but it cannot fully grasp the connection between pixels and the intended element, component, rule, or scope.
For visual modifications, the necessary context is readily available when making the adjustment. You are aware of the target, property, old and new values, viewport width, and whether the change affects a single element or a shared rule. Discarding this context and rephrasing it in English allows your coding agent to attempt a better guess.
Pixy streamlines this process by capturing the visual context of the change on the live site, enabling your coding assistant to better understand and implement the modifications. The agent still retains the authority to decide the implementation method, considering the repository and the specific location of the changed value, whether in Tailwind, a CSS module, a prop, or other global styles.
By describing the change again, you provide your coding agent with another opportunity to accurately interpret your intentions.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
