pdf-prompt-slicer 0.1.1

Slice specific pages from a PDF and copy their text to the clipboard — for building shorter LLM contexts.

pdf-prompt-slicer

A tiny CLI that slices specific pages out of a PDF and copies the extracted text to your clipboard — so you can feed an LLM a focused subset of a document instead of uploading the whole thing.

• Pick pages individually or by range: "[1, 2, 3-6]"
• Get a new, shorter PDF saved next to the original
• Get the raw text of those pages on your clipboard, ready to paste
• Optional clipboard-only mode when you don't need a new PDF

Install (simplest)

Requires uv. If you don't have it:

# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows (PowerShell)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Then install this tool in one line:

uv tool install pdf-prompt-slicer

That puts a pdfslice command on your PATH. You can now run it from anywhere.

Prefer pip? pipx install pdf-prompt-slicer works the same way.

To upgrade later: uv tool upgrade pdf-prompt-slicer To uninstall: uv tool uninstall pdf-prompt-slicer

Usage

pdfslice <path_to_pdf> "<pages>" [-c]

Arguments

• <path_to_pdf> — path to the source PDF.
• <pages> — a list of pages or ranges. Brackets are optional; whitespace is fine. Examples: "[1, 2, 3-6]", "1,2,3-6", "4-4", "7".
• -c, --clipboard-only — copy text to the clipboard only; do not write a sliced PDF.

Examples

Slice pages 1, 2, and 3 through 6 into a new PDF and copy their text:

pdfslice paper.pdf "[1, 2, 3-6]"

Just copy the text (no new PDF):

pdfslice paper.pdf "[1, 2, 3-6]" -c

Output

• The sliced PDF is saved next to the original with _sliced_<n> appended to the filename. paper.pdf → paper_sliced_1.pdf. Running again produces paper_sliced_2.pdf, etc. — existing files are never overwritten.
• The combined text of all selected pages (separated by blank lines) is placed on the system clipboard.

Running without installing

From a clone of the repo:

uv run --script run.py <pdf> "[1, 3-5]"

uv reads the inline script metadata at the top of run.py and sets up a temporary env with the right dependencies — no manual install step.

Development

git clone https://github.com/ChuckySRB/PDFPromptSlicer
cd PDFPromptSlicer
uv sync
uv run pdfslice test_data/your.pdf "[1-3]"

The project layout:

PDFPromptSlicer/
├── pyproject.toml
├── run.py                    # inline-deps entry point (no install needed)
├── src/pdf_prompt_slicer/
│   ├── __init__.py
│   ├── __main__.py           # enables `python -m pdf_prompt_slicer`
│   └── cli.py                # real implementation + `main()`
└── test_data/                # gitignored; for local testing

Dependencies

• pypdf — PDF read/write
• pyperclip — cross-platform clipboard

On Linux, pyperclip needs either xclip or xsel installed for clipboard access. macOS and Windows work out of the box.