AGENT SKILL

extracting-pdf-text

Choose and verify PDF extraction and OCR workflows for text, tables, scans, and complex layouts.

License 许可
Apache-2.0
最近更新
2026-10-12
Stars
★ 1
Forks
0

01 前言 Front Matter

name
extracting-pdf-text
metadata
{ "astralnexus": { "source": "https://www.skills.sh/letta-ai/skills/extracting-pdf-text", "adapter": "curated", "install_command": "npx skills add https://github.com/letta-ai/skills --skill extracting-pdf-text" } }
platforms
[ "windows", "linux", "macos" ]
description
Choose and verify PDF extraction and OCR workflows for text, tables, scans, and complex layouts.

03 SKILL.md 正文

PDF Extraction

Classify the PDF as born-digital, table-heavy, scanned, or complex layout before choosing a parser. Prefer PyMuPDF for ordinary text, pdfplumber for tables, and OCR for scans. Record page, block, table and confidence metadata; validate extracted text against the source page before it becomes evidence.

04 同仓库更多 Skills