Skip to main content

ADX

Agent-native document intelligence layer. Wraps best-in-class document parsers and exposes structured, citeable, inspectable document state to AI agents.

What It Does

ADX is not a parser. It is an orchestration layer that:

  1. Wraps parsers (PyMuPDF, openpyxl, csv) behind a uniform interface
  2. Builds a canonical DocumentGraph from any supported format
  3. Exposes 9 read-only agent tools for inspection and search
  4. Extracts fields using built-in schemas with field-level citations
  5. Validates results with rule-based checks

Install

pip install adx

Quick Start

from adx import ADX

dn = ADX()
doc_id = dn.upload("invoice.pdf")

# Profile the document
profile = dn.profile(doc_id)

# Extract with a built-in schema
extraction = dn.extract(doc_id, schema="invoice")

# Validate
result = dn.validate(doc_id, extraction["id"])

Agent Tools

Tool Description
profile_document File metadata, type detection, recommended tools
list_structure Sections, tables, page outline
search_document Full-text search with citations
get_page Page text blocks and tables
get_table Table rows with headers
list_sheets Spreadsheet sheet metadata
read_range Cell range with formulas
find_cells Search cells by value
inspect_formula Trace formula dependencies

Built-in Schemas

  • invoice — vendor, line items, totals, tax
  • contract — parties, dates, governing law
  • financial_model — revenue, expenses, assumptions
  • table — generic table extraction

REST API

adx serve
curl -X POST http://localhost:8000/v1/files -F "file=@invoice.pdf"
curl http://localhost:8000/v1/files/{id}/profile

CLI

adx upload invoice.pdf
adx profile <id>
adx extract <id> --schema invoice
adx validate <id> --extraction <eid>

Supported Formats

Format Parser Features
PDF PyMuPDF Text blocks, tables, bounding boxes, sections
Excel (.xlsx) openpyxl Sheets, formulas, hidden content, named ranges
DOCX python-docx Paragraphs, headings, tables, images, metadata
RTF striprtf Text extraction with formatting stripped
CSV stdlib Dialect sniffing, encoding detection

Documentation

https://harsh-nod.github.io/adx/

License

MIT

Metadata

Release files for adx 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for adx 0.1.0
File Size Uploaded Provenance Actions
adx-0.1.0.tar.gz 147.9 kB With provenance Details for adx-0.1.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for adx 0.1.0
File Interpreter ABI Platform Actions
adx-0.1.0-py3-none-any.whl Python 3 none any Details for adx-0.1.0-py3-none-any.whl

Total release size: 195.8 kB

Release files / adx-0.1.0.tar.gz

Download URL adx-0.1.0.tar.gz
Size 147.9 kB
Tags Source
SHA-256 checksum
How to use checksums
f9df9b0cd6f673a45410721624f049e654789a7f7543cbf885b386afb1b541f5
BLAKE2b-256 checksum
How to use checksums
0d2e37937aee0f2078a8a6a07a9faea1e27d3e05f77086f8aa82a47b765887ba
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.12

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on May 4, 2026.

Transparency log

Release files / adx-0.1.0-py3-none-any.whl

Download URL adx-0.1.0-py3-none-any.whl
Size 47.9 kB
Tags Python 3
SHA-256 checksum
How to use checksums
c18a843e71b010a8a6078508ba6c35a11c98c78c376b92421841c9ea8f6c494b
BLAKE2b-256 checksum
How to use checksums
658524be7762fbda7ce591783333bcd7608bac0c26a39784dd63f04dde28a846
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.12

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on May 4, 2026.

Transparency log

Security for adx 0.1.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for adx-0.1.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on May 4, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page