run-llama--liteparse
LiteParse — Rust document parser của LlamaIndex: PDF/Word/Excel/PPT/image → JSON với bounding boxes. Node.js/Python/WASM bindings. Tesseract OCR bundled, không cần cloud.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-12
- Stars
- ★ 13
- Forks
- 2
01 前言 Front Matter
- name
- run-llama--liteparse
- description
- LiteParse — Rust document parser của LlamaIndex: PDF/Word/Excel/PPT/image → JSON với bounding boxes. Node.js/Python/WASM bindings. Tesseract OCR bundled, không cần cloud.
- allowed-tools
- Bash, Read, Write
- user-invocable
- true
03 SKILL.md 正文
LiteParse: fast local document parser viết bằng Rust — PDF + office docs + images → structured JSON với spatial bounding boxes. Zero cloud dependency.
Install
# Node.js
npm install liteparse
# Python
pip install liteparse
# Rust
cargo add liteparse
Supported Formats
Native: PDF (trực tiếp, không convert)
Office: .doc .docx .odt .ppt .pptx .odp .xls .xlsx .ods .csv .tsv
Images: JPG PNG GIF BMP TIFF WebP SVG
(qua ImageMagick → OCR)
API
import { parse } from 'liteparse'
// JSON với bounding boxes
const result = await parse('document.pdf', { format: 'json' })
// result.pages[0].blocks[0].bbox = { x, y, width, height }
// result.pages[0].blocks[0].text = "..."
// Plain text với layout preserved
const text = await parse('report.docx', { format: 'text' })
// Screenshot tại DPI tùy chỉnh
await parse('slide.pptx', { format: 'screenshot', dpi: 300 })
from liteparse import parse
result = parse("document.pdf", format="json")
for page in result["pages"]:
for block in page["blocks"]:
print(block["text"], block["bbox"])
Performance
Parallel OCR processing — configurable worker threads
Selective OCR — chỉ chạy khi cần (không phải mọi page)
Page-range targeting — parse trang 10-20, skip phần còn lại
Default DPI: 150 — tùy chỉnh được
OCR Options
Default: Tesseract (bundled, offline)
HTTP backends: EasyOCR, PaddleOCR (accuracy cao hơn)
Custom: implement theo API spec
CLI
liteparse document.pdf --format json --pages 1-10 --output result.json
liteparse batch/ --format text --workers 4
Khi nào dùng
- RAG pipeline: parse docs → chunk → embed
- Extract structured data từ PDF reports
- Convert office docs sang Markdown cho LLM context
- OCR images trong batch pipeline
Source
https://github.com/run-llama/liteparse · Apache-2.0 · +1613⭐/week
04 同仓库更多 Skills
openai--sentry--sentry
Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry API; perform read-only queries with the bundled script and require `SENTRY_AUTH_TOKEN`.
ito-basket-compare
Compare Itô prediction-market baskets against a user's knowledge base, portfolio notes, financial context, watchlist, or research thesis. Use for read-only basket comparison and gap analysis without investment advice or live trading.
implementing-network-traffic-baselining
Build network traffic baselines from NetFlow/IPFIX data using Python pandas for statistical analysis, z-score anomaly detection, and hourly/daily traffic pattern profiling
detecting-evasion-techniques-in-endpoint-logs
Detects defense evasion techniques used by adversaries in endpoint logs including log tampering, timestomping, process injection, and security tool disabling. Use when investigating suspicious endpoint behavior, building detection rules for evasion tactics, or conducting threat hunting for stealthy adversary activity. Activates for requests involving evasion detection, defense evasion analysis, log tampering detection, or MITRE ATT&CK TA0005.
implementing-microsegmentation-with-guardicore
Implementing microsegmentation using Akamai Guardicore Segmentation to map application dependencies, create granular network policies, visualize east-west traffic flows, and enforce least-privilege communication between workloads across data centers and cloud.
ramdisk-tmpfs-patterns
暂无简介