Unpack Word and PowerPoint files to inspect and edit their internal XML, the markup that stores document content. Repack targeted changes with XML syntax checks while preserving the original archive entry order.
中文简介正在准备中。当前显示英语简介。
Low-level support skill for hand-editing the raw XML inside a .docx or .pptx file (both formats are just ZIP archives of XML parts). Use this directly only when you need to unpack, inspect, or repack Open XML yourself — for example fixing a validation error in generated markup, or making one small targeted edit without regenerating a whole document from scratch. The docx and pptx skills already use this internally for their 'edit an existing file' workflow, so you rarely need to load this on its own.
安装前可以查看交给 AI 代理的指令内容。
A .docx or .pptx file is a ZIP archive containing a tree of XML parts (word/document.xml, ppt/slides/slide1.xml, relationship files, content-type manifests, ...). Editing it directly means: unpack the ZIP, edit the XML text, repack the ZIP. No LibreOffice or other converter is required for this — it is pure zipfile + text editing.
Note on language: the docx/pptx/xlsx/pdf skills default to Node (docx/pptxgenjs/exceljs/pdfkit) for creating a file from scratch, per this environment's Node-first policy for office-document generation. This skill stays Python: it's invoked by writing a subprocess call from your sandbox program (whichever coding tool runs it), and the Python invocation (python skills/office-utils/scripts/unpack.py ...) is guaranteed to resolve the same way across every sandbox backend, whereas the TypeScript runtime's binary (tsx) is installed per-sandbox and its exact invocation differs across backends — not worth the portability risk for a script that's about editing XML, not about which language "creates" the file.
| Task | Command |
|---|---|
| Unpack a file to a directory tree | python skills/office-utils/scripts/unpack.py <file.docx|pptx> <out_dir/> |
| Repack a directory tree back to a file | python skills/office-utils/scripts/pack.py <out_dir/> <file.docx|pptx> |
python skills/office-utils/scripts/unpack.py presentation.pptx unpacked/. This extracts every part into unpacked/, pretty-printing every .xml file (one element per line) so edits are readable and diffable. A _unpack_manifest.json file records the original entry order and compression — pack.py reads it back so re-zipping doesn't accidentally change part ordering, which some Office readers are picky about.unpacked/ (e.g. unpacked/word/document.xml, unpacked/ppt/slides/slide1.xml). Prefer the smallest possible edit — a single attribute or run of text — over rewriting a whole XML subtree, so you're less likely to break sibling elements you didn't intend to touch.python skills/office-utils/scripts/pack.py unpacked/ output.pptx. This re-zips using the manifest's original entry order, and validates that every .xml part is still well-formed (parses cleanly) before writing the final file — if any part is malformed, it fails loudly with the offending file name and line instead of silently producing a corrupt document.[Content_Types].xml and _rels/.rels being near the start of the archive. pack.py's manifest-driven ordering exists specifically to avoid this — don't hand-roll your own zipping.r:id) consistent. If you add a new part (e.g. an image), you must add both a <Relationship> entry in the owning part's _rels/*.rels file AND a reference to that same ID from the XML that uses it. Adding one without the other produces a file that "opens" but shows a broken-image icon or corruption warning.&, <, >) in any text you insert into a <w:t>/<a:t> run — a raw ampersand will make the part invalid XML and fail the pack-time validation.Pure Python standard library (zipfile, xml.dom.minidom, json) — no packages to install.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Build or extend a PipesHub connector, an integration that syncs content from an external service or database. The skill guides API research, design decisions, permission mapping, implementation, testing, and documentation of what was actually verified.
中文简介正在准备中。当前显示英语简介。
Load, clean, summarize, and join tabular data from CSV, Excel, JSON, or database exports using pandas. Check missing values, data types, and row counts, then independently verify key figures before reporting numbers printed by the code.
中文简介正在准备中。当前显示英语简介。
Turn data into charts suited to comparisons, trends, distributions, relationships, or shares. Add clear titles, units, and accessible colors, then export a static PNG or interactive HTML with hover details and zoom controls.
中文简介正在准备中。当前显示英语简介。
Create Word reports, memos, letters, and proposals with headings, tables, lists, and a table of contents. Also supports targeted edits to existing .docx files and reading or summarizing their text and tables.
中文简介正在准备中。当前显示英语简介。
Choose suitable libraries for converting spreadsheet data and image files, or extracting text and tables from documents. Covers supported format pairs and identifies conversions unavailable in the described environment before work begins.
中文简介正在准备中。当前显示英语简介。
Create PDF reports, invoices, and letters, extract text or tables from existing PDFs, and fill interactive form fields. The skill also guides table layout, text wrapping, and page breaks so generated documents fit their pages.
中文简介正在准备中。当前显示英语简介。