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 Calix-L/awesome-latex-skills --skill pdf2texgit clone --depth 1 https://github.com/Calix-L/awesome-latex-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/calix-l/awesome-latex-skills/pdf2tex)<a href="https://agentmods.dev/skills/calix-l/awesome-latex-skills/pdf2tex"><img src="https://agentmods.dev/badge/skills/calix-l/awesome-latex-skills/pdf2tex/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/calix-l/awesome-latex-skills/pdf2tex"><img src="https://agentmods.dev/badge/skills/calix-l/awesome-latex-skills/pdf2tex.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 | $0.00046 | $0.01083 |
| Opus 5.5 | $0.00018 | $0.00433 |
| Sonnet 5.5 | $0.00009 | $0.00217 |
| Haiku 4.5 | $0.00005 | $0.00108 |
Grade A, and why
pdf2tex 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 19d 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Establish the reconstruction target
Inspect the PDF, requested pages, available tools, and desired output. Determine whether the user wants content recovery or close visual reconstruction. Preserve the original PDF and write new artifacts to a separate destination.
A PDF may expose text, font names, coordinates, images, and metadata. It does not reliably encode its original document class, packages, macros, bibliography database, comments, or source-file boundaries. Font/creator metadata is evidence for a candidate setup, not proof of the original engine or class.
Extract evidence
When PyMuPDF is available, use the bundled helper from this skill's own directory. The following paths are relative to the repository root; for an installed skill, substitute its actual location. Dependency installation is separate from extraction.
python -m pip install -r pdf2tex/requirements.txt
python pdf2tex/scripts/extract_pdf.py paper.pdf --output extraction --pages 1-3,5 --images --render
The helper creates a new directory with report.html, text.txt, layout.json, and optional
embedded images and whole-page PNG previews when requested. It records page numbers, raw text spans/font/position data,
metadata, selected-page coverage, and warnings. It refuses existing output
directories and refuses publication if the input fingerprint changes during
extraction. Open report.html for offline page/text review; keep the entire
directory together when sharing. It performs no OCR or conversion.
Read PDF extraction guide for API details,
alternative readers, columns, fonts, and OCR. Sorted text is not guaranteed
reading order; inspect page layouts and use coordinates. Images can be repeated
or carry separate soft masks. Vector figures and composite panels often need
a page crop or another export workflow. Use optional --render previews to
inspect selected pages, including vector/composite figures; these are visual
evidence, not OCR or segmented assets. --dpi accepts 72–300 with a per-page pixel limit.
What ships with it
9 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.
- agents/config.yaml 410 B
- agents/openai.yaml 116 B
- assets/review.css 3.9 KB
- references/math-reconstruction.md 6.0 KB
- references/pdf-extraction-guide.md 8.8 KB
- references/structure-detection.md 5.1 KB
- references/table-reconstruction.md 5.6 KB
- requirements.txt 20 B
- scripts/extract_pdf.py 15 KB runs code
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.
- 19d ago Changed 5aae836d7b32
- 21d ago Changed 8e6e6570c3a5
- 22d ago Changed f7bc218e7880
- 23d ago Changed aa3e8ad982cd
- 24d ago Changed · -195 lines · +7 tokens per session 36723e4897b2
- 1mo ago First seen · 289 lines · 39 tokens per session scan A a45fb81767ca
pdf2tex is a skill published in the GitHub repository Calix-L/awesome-latex-skills (184 stars, last pushed 20d ago), licensed MIT. It adds 46 tokens to every session and 1,083 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
overleaf
Sync and manage Overleaf LaTeX projects from the command line. Pull projects locally, push changes back, compile PDFs, and download compile outputs like .bbl files for arXiv submissions. Use when working with LaTeX, Overleaf, academic papers, or arXiv.
fetch-paper
Fetches a legal open-access copy of a paper from its DOI or arXiv ID — resolves through Unpaywall, arXiv PDF/HTML, and the open-access ACM Digital Library. Use when the user wants to fetch, download, read, or get the PDF or full text of a specific paper. Single polite fetch, transient processing only: it never stores…
figure-drawing
MANDATORY skill for creating any figure, diagram, chart, table visualization, or TikZ graphic in a paper. Must be activated BEFORE writing any drawing code. Produces standalone .tex → compiled PDF/PNG → \includegraphics in paper. Never inline TikZ code directly into paper sections.
read-paper-to-notes
Read an academic-paper PDF or supplied full text and turn it into a faithful, structured Markdown deep-reading note. Use when the user provides a research paper and asks to read, explain, summarize, take notes, extract its problem, method, equations, experiments, results, limitations, or research implications, or fill…
user-preview
Compile the current project's LaTeX files and send the generated PDF to the user for preview. Use when the user wants to check typesetting results or confirm the effect of recent edits.
awesome-ieee-report
Generates IEEE-format LaTeX technical reports from project requirements and codebase evidence, using academic prose, pseudocode-only explanations, BibTeX references, and IEEEtran formatting.