/export — Document Export Skill

Running Python helpers

Choose the interpreter before running the examples. For a globally installed copy, use the private runtime recorded by its installer. In a checkout or linked workspace, use AWT_PYTHON when set, otherwise the toolkit's .venv (follow the skill directory link back to the toolkit): Scripts/python.exe on Windows, bin/python on macOS/Linux. Without that environment, check that python (Windows) or python3 (macOS/Linux) actually runs and has the helper's dependencies. Replace the example's python3 with that executable. In PowerShell, prefix a quoted executable with &; keep commands on one line and quote file paths.

Purpose

Convert thesis chapters and reading notes from Markdown to Word (.docx) format and package them into a ZIP archive for submission to supervisors or examiners. This skill is user-invoked only and will not be triggered automatically.

Trigger Words

This skill activates on: /export. It is never invoked automatically by the model.

Parameters

• $ARGUMENTS[0] -- scope: chapters | notes | all (default: all)
• $ARGUMENTS[1] -- language filter: en-only | all (default: all)

Examples:

• /export chapters en-only -- export only chapter files, skip files with significant CJK content
• /export notes all -- export only reading notes, include all languages
• /export -- export everything, all languages

Workflow

0. Resolve every source first (Advisory here; Enforced in the AWT app). Before converting, confirm that every author-year citation in chapters/ has a lint-conforming notes file under literature/reading_notes/, and that references.bib (when present) neither contains entries nothing cites nor lacks an entry a chapter cites:

   python3 .claude/skills/audit/scripts/audit-claim-positioning.py --base-dir chapters --bib references.bib --json
   

   Stop and report if anything is unresolved. On this surface nothing blocks you from exporting anyway — say so if you do. In the AWT app the same condition is a typed denial (EXPORT_SOURCES_UNRESOLVED) on the export_docx tool, and the export does not run.

1. Check dependencies. Verify that pypandoc is installed AND a working pandoc binary is on PATH (the script smoke-probes via pypandoc.get_pandoc_version()). If pypandoc/pandoc are unavailable, fall back to python-docx + markdown. Report which conversion method is being used.

   Run from the project root.

2. Run the conversion script.

   python3 .claude/skills/export/scripts/convert_to_docx.py --base-dir "{project_root}" --output-dir "{project_root}/final_output" --scope {scope} --lang-filter {lang_filter}
   

3. Report results.

   ## Export Complete
   
   | Item | Value |
   |------|-------|
   | Files converted | {N} |
   | Files skipped (language filter) | {N} |
   | Output directory | {path} |
   | ZIP archive | {path} |
   | Total size | {size} |
   
   Conversion method: {pandoc | python-docx fallback}
   

Constraints

1. User-invoked only. This skill has disable-model-invocation: true and must not run unless the user explicitly calls /export.
2. No hardcoded paths. All paths are derived from arguments or project root.
3. No emoji in output.
4. Preserve formatting as much as possible during conversion -- headings, tables, block quotes, inline code.