Word Document Read, Write and Convert

What it does

When your agent uses it

• Summarizing or reading the structure of a Word document, including tracked changes and comments
• Replacing text or inserting images into an existing .docx file
• Converting a Word document to PDF, HTML, Markdown or plain text
• Generating a new Word document such as a report or meeting notes from bullet points

Example prompts

• “Summarize this Word doc and tell me if it has tracked changes.”
• “Replace every instance of 'Company A' with 'Company B' in contract.docx.”
• “Convert this docx to PDF and list its tables.”
• “Generate a sales report document from these bullet points, with a table of contents.”

Requirements

• Node.js with the docx and adm-zip packages for write operations

Workflow steps

4 steps, taken from the step headings in SKILL.md.

1. 1Read: Start with WordInspect(summary)
2. 2Write: JsSandbox + docx
3. 3Format Conversion with DocConvert
4. 4Common Pitfalls

What it can do on your machine

Read from SKILL.md and the folder at commit f7eccf6. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are javascript).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • docx.js.org

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Word Document Read, Write and Convert loads about 1.9k tokens when it runs. Until then it costs about 192 tokens; SKILL.md has 365 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~192

When it runs · the whole SKILL.md, loaded when a task matches

~1.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from OpenLoaf/OpenLoaf at commit f7eccf6, republished under its AGPL-3.0 licence (© OpenLoaf). 365 words, ~1,902 tokens.

name

docx-skill

description

All-in-one Word document (.docx) read / write / convert / review. Trigger scenarios: summarize a docx, read paragraphs / outline / tables / images / comments / tracked changes, replace body text, insert images, change page settings, rebuild table of contents, add comments or replies, add tracked changes (insert/delete/replace), accept/reject revisions, convert docx ↔ pdf/html/md/txt. Typical phrasings: "summarize this Word doc", "change the second paragraph to XXX", "replace all 'Company A' with 'Company B'", "add a tracked change comment", "accept all revisions", "convert docx to pdf", "generate a sales report from these bullet points". Load this skill whenever the user mentions a .docx / .doc file or targets a Word document as the deliverable.

WordInspect { action: "summary", filePath: "…" }

import {
  Document, Packer, Paragraph, TextRun, HeadingLevel, Table, TableRow, TableCell,
  WidthType, AlignmentType, PageOrientation,
} from 'docx'
import fs from 'node:fs/promises'

const title = new Paragraph({
  heading: HeadingLevel.HEADING_1,
  alignment: AlignmentType.CENTER,
  children: [new TextRun({ text: 'Q2 Product Planning — Meeting Notes', bold: true, size: 36 })],
})

const meta = new Paragraph({
  children: [new TextRun({ text: 'Date: 2026-04-20   Attendees: Alice / Bob / Carol', size: 22 })],
})

const decisions = [
  ['Alice', 'Q2 feature schedule', '2026-05-10'],
  ['Bob', 'API documentation', '2026-04-30'],
]
const tbl = new Table({
  width: { size: 100, type: WidthType.PERCENTAGE },
  rows: [
    new TableRow({
      tableHeader: true,
      children: ['Owner', 'Item', 'Due Date'].map(
        t => new TableCell({
          children: [new Paragraph({ children: [new TextRun({ text: t, bold: true })] })],
        }),
      ),
    }),
    ...decisions.map(row => new TableRow({
      children: row.map(v => new TableCell({ children: [new Paragraph(v)] })),
    })),
  ],
})

const doc = new Document({
  creator: 'OpenLoaf',
  styles: {
    default: {
      document: { run: { font: 'Calibri' } },
    },
  },
  sections: [{
    properties: { page: { orientation: PageOrientation.PORTRAIT } },
    children: [title, meta, new Paragraph({ text: '' }), tbl],
  }],
})

await fs.writeFile('meeting_notes.docx', await Packer.toBuffer(doc))
console.log('meeting_notes.docx written')

import AdmZip from 'adm-zip'
import fs from 'node:fs/promises'

const zip = new AdmZip(await fs.readFile('in.docx'))
let xml = zip.readAsText('word/document.xml')
xml = xml
  .replace(/\{\{date\}\}/g, '2026-04-20')
  .replace(/\{\{name\}\}/g, 'John Smith')
zip.updateFile('word/document.xml', Buffer.from(xml, 'utf-8'))
await fs.writeFile('out.docx', zip.toBuffer())
console.log('replaced')

import {
  Document, Packer, Paragraph, ImageRun, Header, AlignmentType,
} from 'docx'
import fs from 'node:fs/promises'

const logo = await fs.readFile('logo.png')

const doc = new Document({
  sections: [{
    headers: {
      default: new Header({
        children: [new Paragraph({
          alignment: AlignmentType.RIGHT,
          children: [new ImageRun({
            data: logo,
            transformation: { width: 80, height: 24 },
          })],
        })],
      }),
    },
    children: [
      new Paragraph({ text: 'Report body...' }),
      new Paragraph({
        children: [new ImageRun({
          data: await fs.readFile('chart.png'),
          transformation: { width: 500, height: 300 },
        })],
      }),
    ],
  }],
})
await fs.writeFile('report.docx', await Packer.toBuffer(doc))

import AdmZip from 'adm-zip'
import fs from 'node:fs/promises'

const zip = new AdmZip(await fs.readFile('in.docx'))
let xml = zip.readAsText('word/document.xml')
// 1) Accept all w:ins (keep content, remove tags)
xml = xml.replace(/<w:ins [^>]*>([\s\S]*?)<\/w:ins>/g, '$1')
// 2) Accept all w:del (remove the deleted content)
xml = xml.replace(/<w:del [^>]*>[\s\S]*?<\/w:del>/g, '')
zip.updateFile('word/document.xml', Buffer.from(xml, 'utf-8'))
await fs.writeFile('out.docx', zip.toBuffer())
console.log('tracked changes accepted')

DocConvert(from="docx", to="pdf",  sourcePath="…")
DocConvert(from="docx", to="md",   sourcePath="…")
DocConvert(from="docx", to="html", sourcePath="…")

© OpenLoaf, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in apps/server/src/ai/builtin-skills/docx/en of OpenLoaf/OpenLoaf.

Open the folder on GitHubat commit f7eccf6