aspiers/ai-config / .agents/skills/agent-browser-viewport
agent-browser-viewport
VerifiedSize the agent-browser Chromium window to fill its current screen, leaving a configurable bottom margin for the desktop panel. Use before any agent-browser automation that needs the full screen visible without elements overlapping or scrolling off (e.g. Xero, dense web apps).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.2 KB |
| scripts/fit-viewport.pyexecutable | script | 3.6 KB |
SKILL.md
View on GitHub---
name: agent-browser-viewport
description: Size the agent-browser Chromium window to fill its current screen, leaving a configurable bottom margin for the desktop panel. Use before any agent-browser automation that needs the full screen visible without elements overlapping or scrolling off (e.g. Xero, dense web apps).
---
# agent-browser viewport sizing
The default `agent-browser` viewport (1280x720) is too short for many
real-world web apps, causing elements to overlap or scroll unexpectedly.
This skill sizes the agent-browser Chromium window to fill its current
screen, leaving a configurable bottom margin for the desktop panel /
status bar.
## Usage
```bash
.agents/skills/agent-browser-viewport/scripts/fit-viewport.py # default 50px bottom margin
.agents/skills/agent-browser-viewport/scripts/fit-viewport.py 80 # larger panel
```
Takes effect immediately on the current agent-browser session — no
reload needed. Run once before starting automation. Re-run if the
window has been moved to a different physical monitor.
## Why a script, not a fixed `-N` fudge
`agent-browser set viewport <w> <h>` sets the **outer** Chromium window
size, but Chromium then adds its tab strip + URL bar on top, so the
actual window ends up larger than the value you pass. The overhead
varies by Chromium version, theme, devicePixelRatio and window manager
— no fixed `-120` constant is reliable across machines.
## How the script works
It derives Chromium's chrome overhead directly from the existing window
state (no probe / resize loop, so the page layout isn't disturbed
mid-fit):
- horizontal overhead = `outerWidth - innerWidth × devicePixelRatio`
- vertical overhead = `outerHeight - innerHeight × devicePixelRatio`
`innerWidth` / `innerHeight` are CSS pixels; `outerWidth` /
`outerHeight` are device pixels. Multiplying inner by DPR brings them
into the same unit, and the difference is the chrome (tab strip + URL
bar + window border) in device pixels — exactly what `set viewport`
needs as its correction.
The script then:
1. Reads the screen the agent-browser window is currently on via
`window-screen-info -n .agent-browser-data` (device pixels). Works
correctly on multi-monitor setups.
2. Reads the in-page `innerWidth` / `innerHeight` / `outerWidth` /
`outerHeight` / `devicePixelRatio` via `agent-browser eval`.
3. Computes overhead as above.
4. Issues one `agent-browser set viewport (screen_width - over_w,
screen_height - bottom_margin - over_h)` call.
The final outer window matches `screen_width × (screen_height -
bottom_margin)` exactly (modulo half-pixel rounding at non-integer
DPR).
## Notes
- The script identifies the agent-browser window by its WM_CLASS
instance `.agent-browser-data` — Chromium sets this from the
`--user-data-dir` path that agent-browser passes.
- Reported screen dimensions are device pixels. `agent-browser set
viewport` also takes device pixels, so values pass through directly.
At `devicePixelRatio > 1` the in-page `innerWidth` / `innerHeight`
will be smaller (CSS pixels = device pixels ÷ DPR), which is
expected.
- If you need a different bottom margin (e.g. taller panel, multi-row
taskbar), pass it as the first argument.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Unverified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |