Extracts text and metadata from PDF files, repairs malformed documents, and produces accessible output.
429 stars
0 votes
0 copies
1 view
Added September 20, 2026
ai-agentsgobash
Security analysis
A100/100
Scanned September 20, 2026
$npx -y skills add smithersai/smithers --skill pdf --agent claude-code
Installs into .claude/skills of the current project.
Are you the author of Pdf?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/smithersai-pdf)
---
name: pdf-processing
description: Extracts text and metadata from PDF files, repairs malformed documents, and produces accessible output.
license: Apache-2.0
allowed-tools: Read Write Bash(pdftotext:*) Bash(pdfinfo:*) Bash(qpdf:*)
metadata:
author: document-tools
version: "1.0"
category: documents
---
# PDF Processing
Use this skill when a task involves inspecting, repairing, extracting, or
rewriting PDF documents.
## Flow
1. Inspect the document with `pdfinfo`.
2. Extract text with `pdftotext -layout` and preserve page boundaries.
3. If the file is malformed, repair it with `qpdf` before retrying extraction.
4. Verify the resulting document opens and that its page count is unchanged.
Do not overwrite the source document unless the user explicitly requests it.