SkillHub

Skills software-mansion argent Argent

argent-screenshot-diff

Use it when:
You want a visual regression test or a before-and-after comparison of a screen around a UI change.
It does:
Saves a full-size baseline screenshot before the interaction or code change, then diffs a fresh capture of the state under test against it.
You get:
A diff summary and artifact files showing what changed in pixels, read together with a normal look at the screen.
Poor fit:
For state, navigation or log checks, finding tap targets, or screens with animation, timestamps, ads or random data.
Needs:
The Argent MCP server.

Skill 6 of 12 in the Argent bundle, after argent-test-ui-flow and before argent-screen-recording.

TL;DR

Part of Software Mansion's Argent toolkit, which lets the agent control simulators, emulators and devices. Argent's screenshot-diff tool highlights pixel-visible change or stability between two screenshots, and this skill sets when and how to use it. It's required when you ask for a visual regression test, a screenshot diff or a before-and-after comparison, unless comparable screenshots can't be produced, and it suits changes in layout, spacing, color, typography, clipping and rendering of text or images. The agent navigates to the known-good state and saves a baseline screenshot at full scale, kept out of its context, before the interaction or edit. Then it makes the change, waits for the app to reload or rebuild, navigates to the state under test and diffs a fresh capture, or a second saved file, against the baseline. The diff is supporting evidence only: the agent still looks at the screen and adds element-tree, log or network checks when the assertion needs them.

Summarizes the skill at b965c94.

npx skills add https://github.com/software-mansion/argent --skill argent-screenshot-diff

skills.sh pageSource on GitHubAll skills from software-mansion/argent

Total Downloads
28,300 on skills.sh, Nov 9, 2026
Repo stars
3,175 software-mansion/argent
First seen
Jun 1, 2026 on skills.sh
Weekly Downloads
1.7K
+26% over the 8 weeks to Nov 9
Weekly values
Week toInstalls
Sep 211,385
Sep 281,528
Oct 51,540
Oct 121,592
Oct 191,643
Oct 261,692
Nov 21,622
Nov 91,745
Category
category:debuggingcategory:computer-usecategory:software-engineering

Find root causes, review code and test what you built.

Drive desktop apps, phones and simulators: tap, type, read the screen and test native apps.

Everything for building software: planning, coding, debugging, design, frameworks, cloud, agents and skills.
Type
type:workflow

A multi-step process the agent follows: interview, debug, run a TDD loop, review.
Effect
effect:local

Its own steps create or change files, git state or local tools.
Needs
needs:mcp

Needs an MCP server connected.
Publisher
by:official

A verified publisher on skills.sh, or the company or maintainer behind the product the skill operates.
Topics
testingverificationmobile
Stack
stack:argent

SKILL.md

From software-mansion/argent at b965c94. Checked daily, last changed Nov 6, 2026.

Frontmatter
name: argent-screenshot-diff
description: Compare saved or live app screenshots with the argent screenshot-diff tool. Use when testing visual regressions, before/after UI comparisons, screenshot diff checks, visible layout, spacing, color, typography, clipping, overflow, text rendering, or image/icon rendering changes.

1. Role

Use screenshot-diff as supporting visual evidence for UI QA and visual regression checks. It highlights pixel-visible change or stability; it does not replace visual inspection, accessibility/component-tree state, frame/attribute checks, logs, network evidence, or app behavior. Its description covers the input combinations, physical-iPhone captures and the result.

Do not use screenshot diffing for tap-coordinate discovery. Use describe or native-describe-screen to find targets first.

2. When To Use

  • Required for explicit "UI regression test", "visual regression test", "screenshot diff", "compare screenshots", or "before/after visual comparison" requests, unless stable comparable screenshots cannot be produced.
  • Good fit when the expected result is pixel-visible, or when the risk is an unintended visual regression outside the exact element changed.
  • Poor fit when the result is better verified structurally: state changes, navigation existence, accessibility tree contents, console/network behavior, or unit tests.
  • Poor fit when dynamic content, unpausable animation, timestamps, ads, random data, or missing baseline/current screenshots would make the comparison noisy or meaningless.

3. Deterministic Flow

  1. Navigate to the known-good state. Capture the baseline before the relevant interaction or before editing whenever feasible.
  2. Save a baseline PNG with screenshot using scale: 1.0 and includeImageInContext: false, so the full-size image stays out of context; keep the returned path. Use normal downscaled screenshots for everything else.
  3. Perform the interaction, apply the code change and navigate to the state under test, after the app reloads or rebuilds.
  4. Call screenshot-diff with that baselinePath, captureCurrent: true, udid, and outputDir. If both images are already saved, pass currentPath instead of captureCurrent.
  5. Inspect the summary and artifact paths, then combine the diff with normal visual inspection and any structural/runtime evidence needed for the assertion.