| name | docling-lab |
| description | Use when the user wants to convert documents with the Docling for IBM watsonx managed service from the command line. Covers converting a single file, a URL, or a directory of documents, choosing output formats, enabling enrichments (OCR, table extraction, picture classification, chart data), and filtering by input format. Trigger phrases: "convert", "docling convert", "convert this document", "convert to markdown", "extract from PDF", "convert with docling", "convert a directory", "docling remote", "docling-lab".
|
Docling Lab — Command-Line Conversion Skill
This skill teaches Bob how to convert documents using the docling convert-remote CLI, which
sends conversion requests to the Docling for IBM watsonx managed service. No local ML models
or GPU are required — all processing happens on the service.
Prerequisites
- Virtual environment from lab section A.1 is active (
.venv)
.env file in lab-1499/ is populated with DOCLING_SERVICE_URL and DOCLING_SERVICE_API_KEYdocling-client is installed (uv pip install docling-client)
Credentials are loaded automatically from the .env file — no need to export them manually.
When to use each reference
| Task | Reference |
|---|
| Convert a file, URL, or directory | convert.md |
| Customize the pipeline (OCR, enrichments, output formats, page range) | options.md |
Always read the relevant reference before running a command.
When the user's request involves a conversion task, activate this skill and route to the
appropriate reference file.
