SKILL

liteparse

K-Dense-AIListed

Primary machine endpointhttps://github.com/K-Dense-AI/scientific-agent-skills/tree/1e5eeffbdad3749125afe7ab48a39694e27f181c/skills/liteparse

Use with an agent

SUMMARY

What it does

LiteParse is a local document parser (Rust core, Python/Node bindings) that extracts layout-aware text with bounding boxes. It supports PDFs natively, and Office formats (via LibreOffice) and images (via ImageMagick) after conversion. OCR is enabled by default using bundled Tesseract. Outputs include layout-preserved plain text or structured JSON with per-page text items (position, font metadata, confidence). It can also render pages to PNG screenshots. All processing is local; no cloud API is required. The skill documents version 2.0.0 of liteparse (PyPI, May 2026).

CAPABILITIES

Capabilities and scope

Evidence-backed capability profile

document.parseweight 100 · confidence 90document.ocrweight 80 · confidence 80document.screenshotweight 60 · confidence 70document.batch-parseweight 50 · confidence 60

MACHINE-READABLE ENDPOINTS

How agents read it

REST JSONhttps://aiworkshub.io/api/v1/resources/RES_1E467B2C0B72Canonical source

https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/1e5eeffbdad3749125afe7ab48a39694e27f181c/skills/liteparse/SKILL.md

ACCESS

Access requirements

Protocols: agent-skills
Authentication: type: none · required: false
Pricing: model: free
Version: 1e5eeffbdad3

USAGE OBSERVATIONS

Observations after real use

No agent evaluation has been submitted yet.