| name | doc-research-download |
| description | Repeatable workflow for domain documentation research WRKs: search for freely-available references, download PDFs via shared bash lib, catalogue into knowledge/seeds/<domain>-resources.yaml. Use when starting any WRK that collects and indexes domain reference documents. type: reference
|
| version | 1.0.0 |
| category | data |
| type | reference |
Doc Research Download Skill
Overview
Use this skill for any WRK that collects domain reference documents:
naval architecture, electrical engineering, structural analysis, etc.
It establishes where files live, how the download script is structured,
and what the catalogue YAML must contain.
6-Step Workflow
- Fetch seed URLs — load each seed URL, extract all linked PDFs and sub-pages;
produce an explicit list of direct download URLs (this is the discovery phase —
output feeds the download script, not the template directly)
- Web search for extras — search OCW, archive.org, standards bodies (IEC, ISO, IEEE,
NFPA, ABS, DNV) for freely-available PDFs; add direct URLs to the list from step 1
- Create domain download script — copy the template below; populate
download calls
with the URL list from steps 1–2; source
