ADE Document Processing Skills plugin
A LandingAI skill: the agent parses PDFs and scans, extracts fields by schema, and splits and classifies documents
- 69 ★ on the repository
- Last commit
Extended access
Why this level
- Document files are sent to the LandingAI cloud for parsing
- The agent installs Python dependencies and runs generated code in your environment
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/ade-document-processing-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Document files are sent to the LandingAI cloud for parsing
- The agent installs Python dependencies and runs generated code in your environment
This is third-party code. Review the repository files before installing.
What it does
The plugin ships one skill, `document-extraction`, for LandingAI Agentic Document Extraction (ADE). It teaches the agent to write Python or TypeScript on top of the REST API and the official libraries: parsing a document into Markdown and hierarchical JSON, extracting fields by a JSON schema or a Pydantic model, splitting a batch into separate documents, classifying pages, building a table of contents and running async jobs for large files. The skill also documents ready pipelines: batch processing, classify first then extract, RAG preparation, stitching tables that span pages, export to CSV and databases, and overlays that draw bounding boxes on the source document. Each value comes back with page coordinates. By default the skill uses the current v2 API and falls back to v1 for the capabilities that live there.
Who it is for. Developers building invoice, statement, form and paper processing who hand the agent the job of writing LandingAI ADE scripts.
Good fit when
- A folder of invoices must become a CSV with line items, descriptions and prices
- Figures from a scientific paper must come out as separate PNG files
- Statements for several months are merged into one transactions table across pages
- A mixed batch of documents is split by type: invoices, receipts, forms
- Documents are prepared for embedding search with semantic chunking
Not a fit when
- You process documents with another service: the skill is written for ADE only
- Documents must not be sent to an external cloud service
- A single plain text PDF where a local library is enough
Example request
Write a script that reads every invoice in ./documents and saves line items, descriptions and prices to CSVLimitations
A `VISION_AGENT_API_KEY` from LandingAI is required and goes into the project `.env`. Documents travel to the LandingAI cloud for parsing, so confidential files call for a look at the service terms first. The free trial is limited and further use is paid. The skill runs nothing itself: the agent writes and executes the code and installs dependencies such as `landingai-ade` into your Python environment.
How to disable. Run `/plugin uninstall ade-document-processing@ade-document-processing-skills` or delete the `document-extraction` folder from `.claude/skills/`. Remove the key from `.env` by hand.
FAQ
How do I install the ADE skill in Claude Code?
Add the marketplace with `/plugin marketplace add landing-ai/ade-document-processing-skills` and install the `ade-document-processing` plugin. By hand, copy the `document-extraction` folder into `.claude/skills/`.
Where do I get VISION_AGENT_API_KEY?
In the LandingAI account settings at va.landing.ai/settings/api-key. The trial period needs no credit card.
Which file formats does the service handle?
The README names more than 20 formats: PDF, images, spreadsheets, presentations and more. No templates or model training are needed.
How is ADE different from plain OCR?
The service parses page structure: tables, forms, columns. Every value arrives with a bounding box and page number, so the result can be checked against the original.
Can the skill handle large files?
Yes, async Parse Jobs and Extract Jobs are documented for long files and heavy extractions.
Discussion
No questions or reviews yet.
Related
PluginsLandingAI's command line parses PDFs and scans into grounded Markdown and extracts schema fields with page-and-box evidence.
A skill that typesets resumes, one-pagers, letters, reports and slides from templates into PDF, PNG and PPTX and checks the result before delivery
PaddleOCR agent skills
PaddleOCR Agent Skills
Two official PaddleOCR skills: text recognition from images and PDFs, and parsing complex documents into Markdown
The agent turns rich answers, files, folders and service exports into an HTML page in a ready style, checked in a browser
Multi-agent documentation generator: analyzes a repository and builds a README and AI assistant config files
A plugin helps agents maintain project documentation, decisions and verification evidence across development sessions
README in short
The page explains what ADE is and lists the requirements: an API key and a Python environment. Two install routes follow, then a table of service strengths and a description of the single skill with its operations and pipelines. Sample agent prompts close the usage section.