Docx conditional sections

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

[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-conditional-sections

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

Handle {{IF_CONDITION}}...{{END_IF_CONDITION}} conditional blocks in Word templates — keep content and strip markers when true, remove entire block when false.

SKILL.md

3.8 KB, as published. Nobody here has run it

Conditional Sections in Word Templates

Pattern

Templates use markers like:

{{IF_RELOCATION}}You are eligible for a relocation package...{{END_IF_RELOCATION}}

Rules:

Condition is true: Keep the content between markers, remove the markers themselves.
Condition is false: Remove the entire block including markers.

Cases to Handle

Conditional blocks may span:

Single paragraph — start and end markers are in the same paragraph
Multiple paragraphs — start marker in one paragraph, end marker in another

Single-Paragraph Conditional

import re

def handle_single_para_conditional(para, condition_key, should_include, data):
    """Handle {{IF_KEY}}...{{END_IF_KEY}} within a single paragraph."""
    start = '{{IF_' + condition_key + '}}'
    end = '{{END_IF_' + condition_key + '}}'

    text = para.text
    if start not in text or end not in text:
        return False  # Not applicable

    if should_include:
        # Strip markers, keep content, then replace placeholders
        inner = re.search(re.escape(start) + r'(.*?)' + re.escape(end), text, re.DOTALL)
        new_text = inner.group(1) if inner else text.replace(start, '').replace(end, '')
        # Replace any remaining placeholders in the content
        new_text = re.sub(r'\{\{([A-Z0-9_]+)\}\}', lambda m: str(data.get(m.group(1), m.group(0))), new_text)
    else:
        new_text = ''

    if para.runs:
        para.runs[0].text = new_text
        for run in para.runs[1:]:
            run.text = ''
    return True

Multi-Paragraph Conditional

When markers span multiple paragraphs, collect and process paragraph-by-paragraph:

from docx.oxml.ns import qn

def remove_paragraph(para):
    """Remove a paragraph element from the document."""
    p = para._element
    p.getparent().remove(p)

def handle_multi_para_conditional(doc, condition_key, should_include, data):
    """Handle conditional block that may span multiple paragraphs."""
    start_marker = '{{IF_' + condition_key + '}}'
    end_marker = '{{END_IF_' + condition_key + '}}'

    paragraphs = list(doc.paragraphs)
    inside = False
    to_remove = []

    for para in paragraphs:
        text = para.text
        starts = start_marker in text
        ends = end_marker in text

        if starts and ends:
            # Entire block in one paragraph
            handle_single_para_conditional(para, condition_key, should_include, data)
        elif starts:
            inside = True
            if not should_include:
                to_remove.append(para)
            else:
                # Strip start marker
                new_text = text.replace(start_marker, '')
                if para.runs:
                    para.runs[0].text = new_text
                    for run in para.runs[1:]: run.text = ''
        elif ends:
            inside = False
            if not should_include:
                to_remove.append(para)
            else:
                new_text = text.replace(end_marker, '')
                if para.runs:
                    para.runs[0].text = new_text
                    for run in para.runs[1:]: run.text = ''
        elif inside:
            if not should_include:
                to_remove.append(para)

    for para in to_remove:
        remove_paragraph(para)

Determining Condition Value

Usually driven by a field in the data:

should_include = data.get('RELOCATION_PACKAGE', '').lower() == 'yes'
handle_multi_para_conditional(doc, 'RELOCATION', should_include, data)

Order of Operations

Always handle conditional sections before general placeholder replacement to avoid replacing placeholders inside blocks that will be removed.