PDF Explore Agent Skill
Use this skill when the user has attached a PDF, paper, report, or other document and the answer needs its content: summarize a section, compare sections, read specific pages, check the table of contents, or read a value off a figure. The `read` tool cannot parse PDF binary — python is the extraction path. Provides `pdf_pages` (pages as text or rendered PNGs, cached) and `pdf_outline` (embedded-bookmark TOC) in the persistent python kernel; load them once via the Kernel Sidecar exec line that `use_skill` appends. For PDF creation/manipulation, use reportlab/pypdf directly.
Install
one command, takes just this skill from the repositorynpx skills add https://github.com/xuzhougeng/wisp-science --skill pdf-exploreThe instruction itself
7 sections, as written by the authorPDF Explore — navigate a PDF without flooding your context
The read tool cannot parse PDFs (binary), and a 50-page PDF pasted
wholesale is ~40K+ tokens. This skill parses the PDF once in the
persistent python kernel (disk + memory cached) so you load only the
pages that matter.
Load first (once per session): run the exec(...) line from the
"Python Kernel Sidecar" section this skill's use_skill output ends
with. Definitions persist across cells; re-run only after a kernel
restart. Requires pypdfium2 (plus pillow for image mode) — if the
first call raises ImportError, install per its hint and re-run.
Which helper
| | when | returns |
|---|---|---|
| pdf_outline(path) | structured doc (paper, report, book) — try this first | [{page, heading, level}, ...] from embedded bookmarks; [] + hint if none |
| pdf_pages(path, pages=[...], mode="text") | the pages/sections you actually need | [{page, text, n_chars}, ...] |
| pdf_pages(path, mode="image", dpi=200, pages=[N]) | figures, scanned pages | PNG per page under .cache/pdf-explore/; view via view_image |
| mode="auto" (default) | unknown PDF | text; flips to image when pages have no text layer (scans) |
Recipe — navigate by outline (try this first)
for e in pdf_outline("paper.pdf"):
print(f"p{e['page']:>3} {' ' * (e['level'] - 1)}{e['heading']}")
Free and instant when the PDF has embedded bookmarks (most
LaTeX-compiled papers do). No LLM fallback in this host: if it returns
[], skim pdf_pages(path, mode="text") first lines per page to build
your own map.
Recipe — read a few pages (≤ ~5)
for p in pdf_pages("paper.pdf", pages=[3, 4, 5], mode="text"):
print(f"\n── page {p['page']} ──\n{p['text']}")
Printing is fine at this scale (~2–4KB/page). Python output beyond the
context budget (~16KB) gets head/tail-truncated at ingestion — so for
anything bigger, use the next recipe instead of printing.
Recipe — pull whole sections for synthesis
For "summarize the methods" / "compare section 3 and 5" / anything
drawing on several page ranges, write the pages to a file in one
call, then read that file — read results enter context whole:
wanted = [5, 21, 22, 23, 24, 25, 62, 63, 64] # from pdf_outline
with open("sections.txt", "w") as f:
for p in pdf_pages("paper.pdf", pages=wanted, mode="text"):
f.write(f"\n── page {p['page']} ──\n{p['text']}")
import os; print(f"wrote {os.path.getsize('sections.txt'):,} bytes")
Then read sections.txt (with offset/limit if it is large).
~800 tokens/page as text vs ~8K tokens as an attached image — and you
pay it once.
Recipe — read a figure in detail
A full page render is too low-res to read axis labels off a dense
figure. Render high-DPI, crop the figure region with PIL, then view the
crop:
p = pdf_pages("paper.pdf", mode="image", pages=[5], dpi=200)[0]
from PIL import Image
Image.open(p["image_path"]).crop((x0, y0, x1, y1)).save("fig_p5.png")
Then call view_image on fig_p5.png (or the full image_path once to
locate the figure). Viewed images persist in context until /compact
ages them — view the few crops that matter, not every page.
Not available in this host
The upstream skill's LLM fan-out helpers (pdf_scan semantic page
ranking, pdf_extract structured sweeps, pdf_map per-page summaries)
need an in-kernel model-call bridge wisp doesn't provide; they were
removed rather than left to NameError. For an exhaustive sweep, dump all
pages to files (recipe above, chunked) and work through them — or
delegate the reading to the explore subagent once the text is on disk.
Other skills for the same job
different authors, same section of the catalogue
Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks
Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.
Presentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.
Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.
Create and edit Obsidian Flavored Markdown with wikilinks, embeds, callouts, properties, and other Obsidian-specific syntax. Use when working with .md files in Obsidian, or when the user mentions wikilinks, callouts, frontmatter, tags, embeds, or Obsidian notes.
How to use it
Take xuzhougeng/pdf-explore from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.