# Redesign and QA PDFs While Keeping Every Detail Intact

Standard PDF editing tools often alter or lose critical information like receipt numbers, dates, amounts, and barcodes. This skill teaches your AI agent to redesign PDFs for readability while preserving every fact, identifier, and machine-readable asset from the source document.

## Install

```bash
npx skillstore add sethupavan12/prettify-pdf
```

## Metadata

- - Slug: sethupavan12-prettify-pdf
- - Version: 1.0.0
- - Author: sethupavan12
- - GitHub username: sethupavan12
- - License: MIT
- - Repository: https://github.com/sethupavan12/prettify-pdf/tree/main/
- - Ref: main
- - Supported tools: Claude, Codex, Claude Code
- - Risk level: safe
- - Risk factors: external\_commands, filesystem
- - Quality score: 86
- - Quality tier: featured
- - Public page: https://skillstore.pages.dev/skills/sethupavan12-prettify-pdf
- - Manifest: https://skillstore.pages.dev/api/skills/sethupavan12-prettify-pdf/manifest

## Capabilities

- Redesign PDF receipts, invoices, forms, tickets, and reports for improved readability
- Preserve exact values of IDs, dates, amounts, names, and other critical fields
- Extract and preserve barcodes, QR codes, seals, signatures, and logos as source images
- Create a preservation checklist of all critical facts before redesigning
- Render final PDF to images and compare visually against the source document
- Flag any clipped text, overlapping elements, or missing machine-readable assets

## Use Cases

- Redesign Payment Receipts for Customer Clarity: An accountant takes a cluttered payment receipt and asks the agent to redesign it so the amount, date, transaction ID, and payer name are immediately visible while preserving the barcode and official seal.
- Process Government Forms for Easier Review: An HR administrator uploads a multi-page government application form and asks the agent to extract the most important fields into a summary card layout while keeping all original data intact.
- QA Redesigned PDFs Against Source Documents: A developer runs a visual comparison between a redesigned invoice and the original, asking the agent to verify that every ID, date, amount, and QR code matches exactly with no clipping or overlap.

## Prompt Templates

### Basic PDF Redesign

```
Use $prettify-pdf to redesign this PDF receipt. Keep all IDs, dates, amounts, and the barcode exactly as they appear in the original.
```

### Structured Redesign with Priority Fields

```
Use $prettify-pdf to redesign this appointment confirmation letter. Put the date, time, location, applicant name, and reference number in a highlighted summary strip at the top. Keep all other fields in clearly labeled sections below.
```

### QA Verification Against Original

```
Use $prettify-pdf to QA this redesigned invoice against the original PDF. Find any changed IDs, totals, dates, payment terms, missing logo, missing barcode, or layout collisions. Report each discrepancy found.
```

### Advanced Custom Layout with Source Preservation

```
Use $prettify-pdf to redesign this multi-page report. Create a one-page executive summary card with title, date, total, and key action items. Keep full detail pages after it. Preserve all original headers, footers, logos, and legal disclaimers exactly. Use green for confirmed totals and amber for deadlines.
```

## Limitations

- Cannot decode or verify the content of barcodes and QR codes
- Scanned PDFs may have incomplete OCR and require manual visual verification
- Does not perform OCR or text recognition on image-only PDFs
- Optional helper script requires PyMuPDF and Pillow packages

## Best Practices

- Always create a preservation list of all IDs, dates, amounts, and machine-readable assets before starting any redesign
- Render the final PDF to images and compare visually against the source to catch layout issues that text extraction misses
- Keep exact original values in their own labeled fields and add friendly interpretations as supplements, never as replacements

## Anti Patterns

- Reformatting or normalizing IDs, timestamps, or amounts can break downstream verification processes - always copy exact source values
- Relabeling ambiguous adjacent text into a guessed relationship changes the document meaning - keep original labels and value groupings
- Applying decorative effects like gradients, low-contrast text, or fake stamps reduces trust and may confuse automated verification systems

## Security Audit

- - Safe to publish: true
- - Audited at: 2026-05-14T05:46:29.648\+00:00
- - Summary: All 474 static analysis findings were evaluated and determined to be false positives. The majority of flags \(285 'Ruby/shell backtick execution', 180\+ 'Weak cryptographic algorithm', 40 'Hardcoded URL', 29 'filesystem'\) are pattern matches against documentation files \(markdown code blocks showing commands, .gitignore entries, GitHub URLs, configuration flags, and text references\). No actual malicious code, obfuscation, credential exfiltration, or network-based data leakage was found. The skill is a legitimate PDF redesign tool with transparent documentation. The only genuine risk factors are filesystem access \(reading/writing PDF files\) and external commands \(invoking a Python helper script\), both of which are necessary and documented features of the skill.

## Stats

- - Views: 0
- - Downloads: 0
- - Favorites: 0
- - Popularity score: 0