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 skills add yuhaoliu7456/pdf2md-skill --skill pdf2md-skillgit clone --depth 1 https://github.com/yuhaoliu7456/pdf2md-skillWrote 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/yuhaoliu7456/pdf2md-skill/pdf2md-skill)<a href="https://agentmods.dev/skills/yuhaoliu7456/pdf2md-skill/pdf2md-skill"><img src="https://agentmods.dev/badge/skills/yuhaoliu7456/pdf2md-skill/pdf2md-skill/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/yuhaoliu7456/pdf2md-skill/pdf2md-skill"><img src="https://agentmods.dev/badge/skills/yuhaoliu7456/pdf2md-skill/pdf2md-skill.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00063 | $0.02964 |
| Opus 5 | $0.00032 | $0.01482 |
| Sonnet 5 | $0.00013 | $0.00593 |
| Haiku 4.5 | $0.00006 | $0.00296 |
Grade A, and why
pdf2-md 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 9d 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 — 327 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PDF-to-Markdown: Academic Paper Conversion
Convert a PDF paper into a single Markdown file by sending pages as images to the LLM and reconstructing the full document.
Strategy Overview
- Render each PDF page to a high-quality image (PNG, 200 DPI)
- Send images to the LLM (yourself) page by page or in batches
- Reconstruct the complete Markdown with correct reading order, headings, figures, tables, and LaTeX equations
Step 0 — Confirm Input
- Ask the user for the PDF path if not already provided.
- Determine output location: default to
<pdf_stem>/content.mdnext to the PDF.
Step 1 — Render PDF Pages to Images
Use pdftoppm (Poppler) to render each page to PNG:
mkdir -p /tmp/pdf2md_pages
pdftoppm -png -r 200 "<pdf_path>" /tmp/pdf2md_pages/page
If pdftoppm is unavailable, use PyMuPDF as fallback:
import fitz
doc = fitz.open(pdf_path)
for i, page in enumerate(doc):
pix = page.get_pixmap(dpi=200)
pix.save(f"/tmp/pdf2md_pages/page-{i+1:03d}.png")
doc.close()
Verify image count matches expected page count.
Step 2 — Page-by-Page LLM Extraction
Read each rendered page image and extract its content. For each page, apply these rules:
2.1 Reading Order
- For two-column layouts: left column first, then right column, top to bottom.
- Full-width elements (title, abstract, full-width figures) at their natural vertical position.
2.2 Heading Hierarchy
CRITICAL: Do NOT assign heading levels by font size alone. Font size only provides initial hints. You MUST determine the final hierarchy by semantic context — the parent-child nesting of sections.
Two-pass approach:
- First pass (per-page): Tag each heading candidate with its visual weight (font size, bold, numbering pattern) and approximate level.
- Second pass (whole-document): After all pages are processed, review the full heading list and correct levels based on these rules:
Rules for heading level assignment:
#— Paper title only (exactly one).##— Top-level sections. Identify these by: (a) they appear in the paper's logical flow as major divisions (Abstract, Introduction, Related Work, Method, Experiments, Conclusion, Acknowledgments, References); (b) they typically share the same font size/weight; (c) numbered sections at the same depth (1, 2, 3…) are all##.###— Subsections that live inside a##section. Identified by sub-numbering (3.1, 3.2) or smaller/different font weight under a parent.####— Sub-subsections (3.1.1) or paragraph-level bold headings.
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.
- 9d ago First seen · 327 lines · 63 tokens per session scan A 4ac6859dc6e0
pdf2-md is a skill published in the GitHub repository yuhaoliu7456/pdf2md-skill (5 stars, last pushed 5mo ago), licensed MIT. It adds 63 tokens to every session and 2,964 once invoked, about $0.0003 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
pydicom
Use pydicom to read, inspect, write, transform, and safely preflight local DICOM datasets and pixel data. Applies to DICOM metadata, transfer syntaxes, compression plugins, frames, private elements, JSON, and bounded de-identification review.
foundry-hosted-agent-validation
Step-by-step process for validating a Python Foundry hosted agent sample (under python/samples/04-hosting/foundry-hosted-agents/) end to end — running it locally (native runtime and azd ai agent run) and after deploying it to an Azure AI Foundry project with azd. Use this when asked to validate a hosted agent sample.
skill-doc-delivery
Convert markdown to DOCX, PPTX, XLSX, PDF office documents — use when you need exportable deliverables.
pdf-extract-create-workflow
Complete PDF lifecycle: download, extract, and generate structured documents with reportlab.
document-direct-python
Use direct Python execution for reliable document creation including spreadsheets, PDFs, and structured reports.
parse-document
Convert a PDF, scan, image of a page, or office file to clean markdown through the connected Superlinked MCP edge, so the source document is not read into model context directly. Use when the user asks to read, parse, OCR, extract from, summarize, or answer questions about a document.