Extracts slide text, tables, and speaker notes from PowerPoint decks into Markdown clippings.
npx skills add https://github.com/ehartye/wiki-master --skill clip-pptx-ehartye
Please help me install this Agent Skill. Skill: clip-pptx Source: https://github.com/ehartye/wiki-master/tree/main/skills/clip-pptx Command: npx skills add https://github.com/ehartye/wiki-master --skill clip-pptx-ehartye
## Slide N heading per slide so citations like "slide 7" stay accurate. - Deduplication and Provenance: Writes clippings with standard frontmatter (source, quality, source-hash), skips duplicates and prior declines, and records thin extractions so empty decks are not retried. - Research Run Tagging: Accepts a --topic flag so all clips from a discovery run can be grouped during triage. - Use Case: During a research run you download a conference talk deck. Run the clipper to store its full text as a Markdown clipping, then hand it to the ingestion workflow for summarizing and cross-referencing into wiki pages. ## Quick Start Ask the agent to clip the file 'reports/q3-strategy.pptx' into the wiki with the topic of your current research run.High-intent search queries and answers about installing and using this skill.
Run the clip-pptx script with the deck path: it uses python-pptx to extract each slide's text, tables, and speaker notes, then writes a Markdown file with one heading per slide. The output is stored as a clipping with source and quality frontmatter.
python-pptx extracts slide text, tables, and speaker notes from .pptx files. Pandoc cannot be used because it has no .pptx input format, and this approach avoids a heavyweight LibreOffice dependency.
No, python-pptx cannot read the binary PowerPoint 97-2003 format. The clipper detects the .ppt extension and fails with a clear message telling you to convert the file to .pptx first.
Windows often maps python3 to a WindowsApps alias stub that fails even when Python is correctly installed. Probe python and py as fallbacks, and install python-pptx into the exact interpreter that will run the helper.
Extraction drops slide design, layout, images, and chart rendering; only table cell text and speaker notes survive. Verify any quotes or figures against the original deck before citing them on a wiki page.
maverickgao8848Generate animation-rich, zero-dependency HTML slide decks from scratch or PPT/PPTX.
akillnessBuild and maintain a persistent markdown wiki from immutable raw sources.
Miosa-osaExtract atomic claims with provenance and confidence from source materials.
OpenSenseNovaExtract text, tables, charts, and image captions from PPTX and PPT presentations.