Docx placeholder replacement

Skill cxcscmu/SkillLearnBench/skills/b1-one-shot-claude-sonnet-4-6/offer-letter-generator/docx-placeholder-replacement

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill docx-placeholder-replacement

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Robustly replace {{PLACEHOLDER}} tokens in Word documents, handling split runs across paragraphs, tables, headers, and footers.

SKILL.md

2.4 KB, as published. Nobody here has run it

Robust Placeholder Replacement in Word Documents

The Split-Run Problem

Word's XML engine often breaks a single word like {{CANDIDATE_NAME}} across multiple runs:

run[0]: '{{CANDI'
run[1]: 'DATE_NAME}}'

A naive run.text.replace() will miss these. Always reconstruct from para.text.

Safe Replacement Algorithm

import re

def replace_in_paragraph(para, data):
    """Replace all {{KEY}} placeholders in paragraph, handling split runs."""
    full_text = para.text
    pattern = re.compile(r'\{\{([A-Z0-9_]+)\}\}')

    if not pattern.search(full_text):
        return  # Nothing to do

    # Build replacement
    def replacer(m):
        key = m.group(1)
        return str(data[key]) if key in data else m.group(0)

    new_text = pattern.sub(replacer, full_text)

    if new_text != full_text and para.runs:
        # Put all text in first run, clear the rest
        para.runs[0].text = new_text
        for run in para.runs[1:]:
            run.text = ''

Apply to Entire Document

def replace_all_placeholders(doc, data):
    """Replace placeholders everywhere: body, tables, headers, footers."""

    # Body paragraphs
    for para in doc.paragraphs:
        replace_in_paragraph(para, data)

    # Tables (with nested table recursion)
    def process_table(table):
        for row in table.rows:
            for cell in row.cells:
                for para in cell.paragraphs:
                    replace_in_paragraph(para, data)
                for nested in cell.tables:
                    process_table(nested)

    for table in doc.tables:
        process_table(table)

    # Headers and footers
    for section in doc.sections:
        for para in section.header.paragraphs:
            replace_in_paragraph(para, data)
        for para in section.footer.paragraphs:
            replace_in_paragraph(para, data)

Notes

Only the first run's formatting is preserved after merging. If parts of the placeholder text have different formatting (e.g., bold {{NAME}}), that formatting is lost — which is usually acceptable for template filling.
Unknown placeholders (keys not in data) are left as-is by default. Adjust the replacer function to raise errors or substitute empty strings if needed.