AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

LiteParse — Local Document & PDF Parsing

Extract layout-preserved text with per-item bounding boxes, run OCR, and render page PNGs from PDFs, Office files, and images — all fully local.

Docs & OfficeIntermediate★ 33,030⑂ 3,248AI score 9/10Last updated: Aug 9, 2026

What it does

  • Extracts layout-preserved plain text from PDFs, DOCX/XLSX/PPTX, and image files.
  • Produces structured JSON with per-page text_items including position (x, y, width, height), font metadata, and confidence — ideal for layout-aware RAG and citation grounding.
  • Runs OCR on scans with bundled Tesseract; configurable language, DPI, worker count, or an external HTTP OCR server (air-gapped friendly via TESSDATA_PREFIX).
  • Renders page screenshots (PNG) so multimodal agents can see figures, dense tables, and handwriting.
  • Supports batch directory ingestion, page subsets, password-protected PDFs, and stdin/bytes input.
  • Includes a decision table for when to use markitdown, the pdf skill, or LlamaParse instead.

Who it's for

  • Engineers and researchers building RAG or literature-ingestion pipelines.
  • Teams that need source-region highlighting or coordinate-based citation grounding.
  • Anyone required to keep documents on-premises with no cloud API calls.

Examples

  1. lit parse paper.pdf --format json -o paper.json, then use the bboxes to highlight the exact quoted sentence in a PDF viewer.
  2. Bulk-preprocess a literature folder: lit batch-parse ./papers ./parsed --format json --recursive.
  3. For chart-heavy decks: lit screenshot deck.pdf --dpi 300 -o ./shots and feed the PNGs plus JSON text to a multimodal model.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/HEAD/skills/liteparse/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the skills/liteparse folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/liteparse/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git /tmp/sas && mkdir -p ~/.claude/skills && cp -r /tmp/sas/skills/liteparse ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and clone the repo: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  2. Copy the skill into Claude's skills directory: mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/liteparse ~/.claude/skills/
  3. Confirm Python 3.10+: python --version
  4. Install the parser: uv pip install liteparse (if the pinned ==2.0.0 version from the docs isn't on PyPI yet, install the latest release).
  5. Verify: lit --help and python -c "import liteparse; print(liteparse.__version__)"
  6. Optional: install LibreOffice for Office formats and ImageMagick for images; make sure soffice and magick/convert are on your PATH.
  7. Restart Claude Code and try a prompt like "parse this PDF into JSON with bounding boxes" to trigger the skill.
View source on GitHub ↗License: MIT