Menu
Robustly extracts text from various document formats with multiple fallback options.
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/cxcscmu/SkillLearnBench --skill run2-robust-text-extractionThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Showing SKILL.md
| name | run2_robust_text_extraction |
| description | Robustly extracts text from various document formats with multiple fallback options. |
Extract text from PDF, DOCX, and PPTX even when specialized tools are missing.
Primary: pdftotext -l 5 <file> - (extracts first 5 pages).
Fallback: strings <file> | head -n 100 (last resort to find metadata/titles).
Primary: pandoc <file> -t plain.
Fallback (Python):
import zipfile, xml.etree.ElementTree as ET
def get_docx_text(path):
with zipfile.ZipFile(path) as z:
xml_content = z.read('word/document.xml')
tree = ET.fromstring(xml_content)
return ' '.join(node.text for node in tree.iter() if node.text)
Primary: python3 -m markitdown <file>.
Fallback (Python):
import zipfile, xml.etree.ElementTree as ET
def get_pptx_text(path):
text = []
with zipfile.ZipFile(path) as z:
for f in sorted(z.namelist()):
if f.startswith('ppt/slides/slide'):
xml_content = z.read(f)
tree = ET.fromstring(xml_content)
text.append(' '.join(node.text for node in tree.iter() if node.text))
return ' '.join(text)
Based on SOC occupation classification