Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Before And After

ASecurity

Capture before/after screenshots of pages, elements, or images. Use when UI changes need visual comparison or pull-request evidence.

2 stars
0 votes
0 copies
0 views
Added 9/23/2026
ai-agentsbashnodegit

Works with

cli

Security Analysis

A92/100
mediumInstalls packages at runtime which could introduce malicious dependencies

Scanned 9/23/2026

Install to Claude Code

$npx -y skills add jaydubya818/MissionControl --skill before-and-after --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Before And After?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Before And After
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jaydubya818-before-and-after/badge)](https://www.skillsdirectory.com/skills/jaydubya818-before-and-after)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: before-and-after
description: Capture before/after screenshots of pages, elements, or images. Use when UI changes need visual comparison or pull-request evidence.
license: LICENSE
compatibility: Requires Node.js for capture. Upload and PR integration require network access and authenticated provider tooling.
metadata:
  author: vercel-labs
  version: "2.0.0"
  owner: software-factory
  risk: medium
  capabilities: visual-evidence,pr-evidence
allowed-tools: >-
  Bash(npx @vercel/before-and-after *) Bash(before-and-after *)
  Bash(which before-and-after) Bash(*/upload-and-copy.sh *)
  Bash(curl -s -o /dev/null -w *) Bash(gh pr view *) Bash(gh pr edit *)
  Bash(vercel inspect *) Bash(vercel whoami) Bash(which vercel) Bash(which gh)
---

# Before-After Screenshot Skill

> **Package:** `@vercel/before-and-after`
> Never use `before-and-after` (wrong package).

## Agent Behavior Rules

**DO NOT:**
- Switch branches or stash another task's changes. Use an isolated worktree for
  any baseline server already authorized by the task.
- Use `--full` unless user explicitly asks for full page / full scroll capture

**DO:**
- Use `--markdown` when user wants PR integration or markdown output
- Use `--mobile` / `--tablet` if user mentions phone, mobile, tablet, responsive, etc.
- Reuse before captures, baseline URLs, and revision information already
  established in the task. Confirm that they show the relevant behavior and
  match the after capture's viewport and starting state.
- Treat the current state as after only when the task context establishes it.
  If the before state remains ambiguous after checking available context, ask
  for the missing baseline. Do not ask again for a URL when a valid image pair
  or earlier baseline is already available.

## Capture and delivery

1. Use an installed `before-and-after` CLI or `npx @vercel/before-and-after`.
2. For remote URLs, check access when needed. A 401/403 indicates protection.
   Existing image pairs do not require a URL or deployment access check.
3. Capture missing images with `before-and-after "<before-url>" "<after-url>"`.
4. When the task calls for uploaded evidence or PR markdown, run
   `before-and-after before.png after.png --markdown` once. For local capture
   requests, deliver the files without uploading them.
5. Append the markdown to the PR when PR integration is part of the task.

Resolve bundled script paths relative to this skill's directory, not the
project working directory. Reuse earlier evidence instead of recapturing it.

## Quick Reference

```bash
# Basic usage
before-and-after <before-url> <after-url>

# With selector
before-and-after url1 url2 ".hero-section"

# Different selectors for each
before-and-after url1 url2 ".old-card" ".new-card"

# Viewports
before-and-after url1 url2 --mobile    # 375x812
before-and-after url1 url2 --tablet    # 768x1024
before-and-after url1 url2 --full      # full scroll

# From existing images
before-and-after before.png after.png --markdown

# Via npx (use full package name!)
npx @vercel/before-and-after url1 url2
```

| Flag | Description |
|------|-------------|
| `-m, --mobile` | Mobile viewport (375x812) |
| `-t, --tablet` | Tablet viewport (768x1024) |
| `--size <WxH>` | Custom viewport |
| `-f, --full` | Full scrollable page |
| `-s, --selector` | CSS selector to capture |
| `-o, --output` | Output directory (default: ~/Downloads) |
| `--markdown` | Upload images & output markdown table |
| `--upload-url <url>` | Custom upload endpoint (default: 0x0.st) |

## Image Upload

```bash
# Default (0x0.st - no signup needed)
./scripts/upload-and-copy.sh before.png after.png --markdown

# GitHub Gist
IMAGE_ADAPTER=gist ./scripts/upload-and-copy.sh before.png after.png --markdown
```

## Vercel Deployment Protection

If `.vercel.app` URL returns 401/403:

1. Check Vercel CLI: `which vercel && vercel whoami`
2. If available: `vercel inspect <url>` to get bypass token
3. If not: Tell user to provide bypass token, take manual screenshots, or disable protection

## PR Integration

```bash
# Check for gh CLI
which gh

# Get current PR
gh pr view --json number,body

# Write the existing body plus the comparison to a temporary file, then:
gh pr edit <number> --body-file /path/to/pr-body.md
```

If no `gh` CLI: output markdown and tell user to paste manually.

## Error Reference

| Error | Fix |
|-------|-----|
| `command not found` | `npm install -g @vercel/before-and-after` |
| `could not determine executable` | Use `npx @vercel/before-and-after` (full name) |
| 401/403 on .vercel.app | See Vercel protection section |
| Element not found | Verify selector exists on page |

Attribution

jaydubya818jaydubya818
View sourceMore from jaydubya818 →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

693621 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →