docx

Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.

1.52x

Quality

90%

Does it follow best practices?

Impact

93%

1.52x

Average score across 8 eval scenarios

Securityby

Passed

No findings from the security scan

The canonical home for this skill is docx in anthropics/skills

Quality

Content

80%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, concise skill body with executable code and clear gotchas; its main weaknesses are the editing workflow's implicit (rather than explicit) validation feedback loop and a monolithic structure that keeps reference material inline instead of splitting it out.

Suggestions

Add an explicit validation checkpoint to the editing workflow (e.g. Step 2.5: run `python scripts/office/validate.py` after editing XML, fix and re-validate, then pack) so the validate→fix→retry loop is explicit rather than bundled into pack.

Consider moving the XML Reference and Critical Rules for docx-js into separate reference files (e.g. XML_REFERENCE.md) referenced one level deep from SKILL.md to reduce the monolithic size and aid discovery.

Trim the restated page-size/width rules that duplicate earlier guidance to tighten the token budget further.

Dimension	Reasoning	Score
Conciseness	The body is lean and action-dense — executable bash/JS/XML snippets, terse one-line overview ("A .docx file is a ZIP archive containing XML files"), and consolidated gotcha lists rather than tutorials explaining concepts Claude already knows; minor restatement of page-size rules is intentional consolidation.	3 / 3
Actionability	Throughout it provides fully executable, copy-paste-ready commands and code with exact paths (e.g. `python scripts/office/unpack.py`, complete docx-js snippets, concrete tracked-change XML), not vague direction or pseudocode.	3 / 3
Workflow Clarity	The editing flow is explicitly sequenced ("Follow all 3 steps in order": Unpack → Edit → Pack) and validation exists inside pack with auto-repair details, but the editing path lacks an explicit standalone validate→fix→re-validate checkpoint before packing, unlike the creation path's explicit feedback loop.	2 / 3
Progressive Disclosure	It is well-organized into clear sections and all referenced scripts exist, but everything lives in a single ~580-line file with no external reference docs; the XML Reference and Critical Rules sections are inline content that could reasonably be split out into one-level-deep reference files.	2 / 3
	Total	10 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: it names concrete capabilities, gives rich natural-language triggers covering both what and when, and draws a clear boundary against adjacent skills with an explicit exclusion clause.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions such as "create, read, edit, or manipulate", "extracting or reorganizing content", "inserting or replacing images", "find-and-replace", "tracked changes or comments", and "converting content into a polished Word document", matching the multiple-specific-actions anchor.	3 / 3
Completeness	Explicitly answers both what (create/read/edit/manipulate Word documents and listed sub-tasks) and when via explicit "Use this skill whenever...", "Triggers include", "Also use when", and "If the user asks for..." clauses.	3 / 3
Trigger Term Quality	Covers natural terms users would actually say — "Word doc", "word document", ".docx", "report", "memo", "letter", "template", plus formatting cues like "tables of contents, headings, page numbers, or letterheads".	3 / 3
Distinctiveness Conflict Risk	Clear Word/.docx niche with distinct triggers and an explicit exclusion ("Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks"), making conflicts with other skills unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (591 lines); consider splitting into references/ and linking	Warning
referenced_paths_exist	Referenced path issues: 7 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: tsinghua-fib-lab/AgentSociety
Path: extension/skills/docx/SKILL.md
Commit: 1832bcb

Reviewed: about 5 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.