# Agentic Document Extraction (ADE)

Python library for Agentic Document Extraction (ADE).

Repository: https://github.com/landing-ai/ade-python
Canonical: https://ross.abutalabs.com/products/agentic-document-extraction-ade
Homepage: https://ade.landing.ai
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-25T04:16:42+00:00

## Health v2 (maintenance only)
Score: 84/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 98, longevity 24
- inputs: {"age_days": 349, "days_push": 8, "days_rel": 12, "gap_med": 0.5, "n_releases_24m": 59}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1028, forks 167 (observed 2026-08-28T04:03:17.447310+00:00)

## What it is
The official CLI for LandingAI's Agentic Document Extraction (ADE), which parses documents into grounded Markdown and elements and extracts schema-shaped fields with page-and-box evidence. It ships self-contained binaries for macOS, Linux, and Windows and caches results locally at ~/.ade to avoid redundant API credit usage.

## Use cases
- parse pdf invoices into markdown with grounded elements
- extract structured fields from documents using a json schema
- convert scanned documents with tables and charts to markdown
- find where an extracted value came from in the source document
- crop regions of a parsed document page as images
- search parsed document elements locally without api calls

## When to choose
- you need schema-shaped structured data extracted from documents with per-field evidence
- you want a terminal-first workflow for document parsing without installing Python
- you need grounded extraction that shows page-and-box provenance for each value
- you process documents repeatedly and want local caching to save API credits

## When to avoid
- you need fully offline document extraction since ADE requires a LandingAI API key and cloud service
- you want a free open-source OCR engine rather than a paid API client
- you need batch processing pipelines better served by the ade-python SDK
- you require self-hosted document processing

## Facets
- artifact type: cli-tool
- maturity: active
- function: ocr, parser, pdf, llm-inference, cli
- domain: pdf, developer-tools, artificial-intelligence
- platform: windows, cli, python
- tags: document-extraction, agentic-document-extraction, schema-extraction, landingai, document-parsing, grounded-markdown, automation, macos, linux

## Member repositories
- landing-ai/ade-python (main) score 84
- landing-ai/ade-cli (plugin) score 85

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:17.447310+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:34:07.144682+00:00, confidence not recorded.
  - readme: https://github.com/landing-ai/ade-python (fetched 2026-08-28T04:03:17.447310+00:00, sha 46b14de04b5e)
  - registry_pypi: https://pypi.org/pypi/ade-python/json (fetched 2026-08-29T10:14:22.481966+00:00, sha 47c251d9b8e4)
- Data as of 2026-08-30T08:39:29.467469+00:00.