Kordoc parse form
Use when extracting structured label-value fields from Korean forms with `npx` and kordoc. Trigger for requests involving applications, government forms, personnel sheets, contact fields, or any HWP, HWPX, or PDF document that should be processed with `kordoc:parse_form`.From its SKILL.md
npx -y skills add composite/korean-skills --skill kordoc-parse-formAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
- runs commandsInstructs the agent to run 1 command, including `npm exec --yes --package=kordoc --package=pdfjs-dist -- node <<'SCRIPT' ...`.
SKILL.md
1.9 KB, 393 tokens by cl100k_base, as published. Nobody here has run it
kordoc: parse_form
Reproduce the parse_form logic from src/mcp.ts: parse the document, then run extractFormFields.
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. - Prefer form-like or table-heavy office documents.
- Do not proceed if the document is missing.
Command
npm exec --yes --package=kordoc --package=pdfjs-dist -- node <<'SCRIPT'
const p = require("path"), fs = require("fs");
const nmBin = process.env.PATH.split(p.delimiter).find(d => d.endsWith(".bin"));
const kUrl = "file:///" + p.join(p.resolve(nmBin, ".."), "kordoc", "dist", "index.js").split(p.sep).join("/");
import(kUrl).then(async (k) => {
const filePath = "/abs/path/form.hwpx";
const raw = fs.readFileSync(filePath);
const buffer = raw.buffer.slice(raw.byteOffset, raw.byteOffset + raw.byteLength);
const result = await k.parse(buffer);
if (!result.success) throw new Error(result.error);
console.log(JSON.stringify(k.extractFormFields(result.blocks), null, 2));
}).catch(e => { console.error(e); process.exit(1); });
SCRIPT
Workflow
- Resolve the document path to an absolute path.
- Parse the file once.
- Run
extractFormFields(result.blocks). - Return structured JSON and preserve confidence if present.
Guardrails
- Keep the output structured.
- Do not invent missing field values.
- If the document is not form-like, say so and suggest a normal parse instead.
- Refuse to continue when no supported document was provided.
What ships with it: 1 file
217 B alongside SKILL.md
agents/
- openai.yaml217 B