| name | kordoc-parse-pages |
| description | Use when parsing only part of a Korean document with `npx` and kordoc. Trigger for requests such as extracting pages 1-3, reading selected pages from HWP, HWPX, or PDF, limiting work on large files, or running `kordoc:parse_pages` with a page-range string. |
kordoc: parse_pages
This skill maps directly to the parse_pages intent in src/mcp.ts. Use the public CLI.
Input Requirements
- Require exactly one target document.
- Require the user to attach the document or provide a concrete local path.
- Accept only supported formats:
.hwp, .hwpx, .pdf.
- Require an explicit page range such as
1-3 or 1,3,5-7.
- Do not proceed if the document or page range is missing.
Commands
npm exec --yes --package=kordoc --package=pdfjs-dist -- \
kordoc /abs/path/document.hwpx --pages 1-3
npm exec --yes --package=kordoc --package=pdfjs-dist -- \
kordoc /abs/path/document.hwpx --pages 1,3,5-7 --format json --silent
Workflow
- Resolve the document path to an absolute path.
- Normalize the requested range such as
1-3 or 1,3,5-7.
- Use plain output for reading tasks.
- Use JSON when the caller needs structured blocks or metadata.
- Repeat the exact requested range in the answer.
Guardrails
- PDF page selection is exact.
- HWP and HWPX page slicing is section-based and approximate.
- If the user asks for one page from a non-PDF document, warn about that approximation.
- Refuse to continue when no supported document was provided.
