菜单
Unified multi-format extraction for PDF, DOCX, and PPTX with consistent output and quality metrics
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/cxcscmu/SkillLearnBench --skill run2-multi-format-extraction命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Handles reading, populating, and saving .docx files using the python-docx library. Use this skill for any tasks involving template filling or modifying Word documents.
Perform various data analysis on SEC 13-F and obtain some insights of fund activities such as number of holdings, AUM, and change of holdings between two quarters.
This skill includes search capability in 13F, such as fuzzy search a fund information using possibly inaccurate name, or fuzzy search a stock cusip info using its name.
基于 SOC 职业分类
正在显示 SKILL.md
| name | run2_multi-format-extraction |
| description | Unified multi-format extraction for PDF, DOCX, and PPTX with consistent output and quality metrics |
Unified framework for extracting text from multiple document formats with quality metrics and consistent output. Handles edge cases and provides confidence scores for extraction quality.
pip install pdfplumber python-docx python-pptx PyPDF2 -q
from pathlib import Path
from dataclasses import dataclass
import pdfplumber
from docx import Document
from pptx import Presentation
@dataclass
class ExtractionResult:
"""Container for extraction results with quality metrics"""
text: str
format: str
length: int
quality_score: float # 0.0-1.0
success: bool
def extract_document_unified(file_path, max_content_units=5):
"""
Unified extraction for PDF, DOCX, PPTX.
Returns ExtractionResult with quality score.
"""
file_path = Path(file_path)
ext = file_path.suffix.lower()
if ext == '.pdf':
return extract_pdf_robust(file_path, max_content_units)
elif ext == '.docx':
return extract_docx_robust(file_path, max_content_units)
elif ext == '.pptx':
return extract_pptx_robust(file_path, max_content_units)
else:
return ExtractionResult("", "unknown", 0, 0.0, False)
def extract_pdf_robust(pdf_path, max_pages=5):
"""Extract from PDF with quality scoring"""
text = ""
try:
with pdfplumber.open(pdf_path) as pdf:
for page in pdf.pages[:max_pages]:
extracted = page.extract_text()
if extracted:
text += extracted + " "
except Exception as e:
pass
quality = min(1.0, len(text) / 1000.0) # 1000 chars = perfect quality
return ExtractionResult(text, "pdf", len(text), quality, len(text) > 50)
def extract_docx_robust(docx_path, max_paragraphs=30):
"""Extract from DOCX with quality scoring"""
text = ""
try:
doc = Document(docx_path)
for para in doc.paragraphs[:max_paragraphs]:
if para.text.strip():
text += para.text + " "
except Exception as e:
pass
quality = min(1.0, len(text) / 1000.0)
return ExtractionResult(text, "docx", len(text), quality, len(text) > 50)
def extract_pptx_robust(pptx_path, max_slides=5):
"""Extract from PPTX with quality scoring"""
text = ""
try:
prs = Presentation(pptx_path)
for slide in prs.slides[:max_slides]:
for shape in slide.shapes:
if hasattr(shape, "text") and shape.text.strip():
text += shape.text + " "
except Exception as e:
pass
quality = min(1.0, len(text) / 1000.0)
return ExtractionResult(text, "pptx", len(text), quality, len(text) > 50)
result = extract_document_unified(file_path)
if result.quality_score > 0.3: # Reasonable confidence
category = classify_text(result.text)
else:
category = 'music_history' # Safe default