Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/alexagriffith/agent-skills/html2decknpx skills add alexagriffith/agent-skills --skill html2deckgit clone --depth 1 https://github.com/alexagriffith/agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/alexagriffith/agent-skills/html2deck)<a href="https://agentmods.dev/skills/alexagriffith/agent-skills/html2deck"><img src="https://agentmods.dev/badge/skills/alexagriffith/agent-skills/html2deck.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00079 | $0.02859 |
| Opus 5 | $0.00039 | $0.01430 |
| Sonnet 5 | $0.00016 | $0.00572 |
| Haiku 4.5 | $0.00008 | $0.00286 |
Grade A, and why
html2deck scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 244 lines — stays where its author put it; the contents beside it link to each section on GitHub.
html2deck
What it does
Given an HTML page, html2deck renders it in headless Chromium at exactly
1280x720 (16:9), walks the page's top-level "units" (each --unit-selector
match), projects a slide face (visual-first; excess prose → speaker notes),
and decides — per unit — whether it fits on one slide or must be split.
It then writes one PNG per slide, a contact sheet for approval, and finally a
.pptx and .pdf.
Fixed title bar (PowerPoint-master style)
The heading is pulled OUT of the flowing content into a pinned top zone — same
Y and same font size on EVERY content slide, like a PowerPoint master. Only the
content below it (figure / table / lead) is scaled to fit; the title never moves
or resizes from slide to slide. On a split unit the N / M part label sits
top-right of that title bar.
Fit equation (hard — never overflow):
[ s_{\mathrm{fit}}=\min!\Big(1,\ \tfrac{H_c}{h},\ \tfrac{W_c}{w}\Big),\quad s=\begin{cases} s_{\mathrm{fit}} & h>H_c\ \min(s_{\max},,0.72,H_c/h) & \text{visual, under-filled}\ 1 & \text{otherwise} \end{cases} ]
Readability floors (0.90 / 0.80) decide whether to split, never whether to
spill. Content is clipped in #__fit under the title; the title zone is
z-stacked with an opaque background so transforms cannot paint over it.
Content stays top-aligned under the title.
Title page exception: the first unit's first slide uses a PowerPoint title layout — title + optional lead stacked and centered on the full canvas (no pinned title bar). Never a graphic, table, card, or figure on a title page. If the opener needs a visual, the title page stays text-only and the visual lands on the next slide (master title bar), or under the next section header. The same title-page layout may be reused as a section header before a graphic-heavy unit — title alone (or title + one lead), then the chart/table.
Layout contract: best-fit equation
The height budget alone prevents overflow. Composition fit prevents awkward slides (sparse, text-walled, or chart-crushed-by-prose).
What ships with it
19 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- .gitignore 123 B
- build.py 3.6 KB runs code
- contact_sheet.py 3.6 KB runs code
- examples/01-title-slide.png 50 KB
- examples/02-chart-slide.png 111 KB
- examples/03-text-only-centered.png 72 KB
- examples/04-card-grid.png 151 KB
- examples/05-table-and-diagram.png 150 KB
- examples/06-pipeline-strip.png 73 KB
- examples/07-homepage-export-menu.png 108 KB
- examples/07-homepage-export.png 99 KB
- grade_slides.py 6.3 KB runs code
- layout.py 17 KB runs code
- README.md 3.8 KB
- scripts/bootstrap.sh 293 B runs code
- slice.py 55 KB runs code
- tests/fixtures/sample.html 3.2 KB
- tests/test_slice.py 16 KB runs code
- TODO.md 1004 B
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 244 lines · 79 tokens per session scan A d69eb50af9b2
html2deck is a skill published in the GitHub repository alexagriffith/agent-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 79 tokens to every session and 2,859 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
instrument-data-to-allotrope
Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format or flattened 2D CSV. Use this skill when scientists need to standardize instrument data for LIMS systems, data lakes, or downstream analysis. Supports auto-detection of instrument types. Outputs include full…
baoyu-youtube-transcript
Downloads YouTube video transcripts/subtitles and cover images by URL or video ID. Supports multiple languages, translation, chapters, and speaker identification. Caches raw data for fast re-formatting. Use when user asks to "get YouTube transcript", "download subtitles", "get captions", "YouTube字幕", "YouTube封面"…
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
gog-slides
Google Slides operations through gog.
read
Reads URLs and PDFs by fetching source content, defaulting to concise summaries for plain read requests and clean Markdown when asked to convert, save, quote, cite, or feed downstream work. Use when users ask in any language to read, fetch, check, summarize, quote, cite, convert, or save a URL or PDF. Not for local…