Extract a PowerPoint (.pptx) file into the user's plain-text slide authoring format: a Markdown title + body per slide, presenter instructions in [brackets], a `===` separator, then speaker notes, with `---` between slides. Use this whenever the user uploads or points to a .pptx and wants the slides "as text", "in my slide format", reverse-engineered back into editable source, or wants the bullets / notes / animations / images pulled out of an existing deck. Trigger even if they just say "turn t
SkillScore86Maintained
Automatic
◆Doc Ops pack
Last verified 8/9/2026
What it changes
Passed 16 of 19 test cases with the skill, and 7 of 19 without it.
Without skill36%
With skill86%
Checks it passed that the model alone missed
Flags hidden slides with `[hidden slide]`
Specifies `--images-dir assets/extracted_imgs/`
Formats charts as `[chart: <type> — "<title>"]`
and 9 more in the full report
11 fixed8 passed either way3 still failing0 broke
Pass-rate lift
+50 pp
Tokens
+53%
Agent turns
0%
Response time
-69%
Model gemini-3.6-flashMeasured 8/9/2026v219 of 22 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 11 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 11
Prompt
When extracting slide 2 containing an image with alt text 'Revenue chart for 2023' saved to `images/slide2_img1.png`, what exact inline bracketed string is placed in the slide content area?
What was checked, without and with the skill
Checked
Without skill
With skill
Specifies `[image: images/slide2_img1.png — "Revenue chart for 2023"]`
Missed
Passed
case-13 · gemini-3.6-flash · 8/9/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md162 lines2 filesincludes scripts
PPTX → slide-Markdown
Convert an existing .pptx back into the plain-text format the user writes
slides in, so a finished deck becomes editable source again.
Target format
One block per slide. --- separates slides; === separates slide content from
speaker notes. Presenter instructions (images, animations, demos) live in
[brackets] inside the content area. If the deck has sections, each section
begins with a [Section N: Section Name] label before the first slide in that
section.
[Section 1: Introduction]
[Slide 1]
# Slide Title
Slide content in markdown: bullets, numbered lists, tables, **bold**, *italic*, ~~strikethrough~~, [links](url).
[image: assets/slide3_img1.png — "alt text"]
[animation — 2 click steps:
click 1: entrance of "First point"
click 2: entrance of "Second point"]
===
**Speaker Notes:**
Notes in markdown.
---
[Slide 2]
# Next Slide Title