word-read-write

Create and read Microsoft Word (.docx) documents. Use this skill when you need to generate reports/letters/templates as .docx or extract readable text from existing .docx files.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Other/word-read-write/SKILL.md

Quality

Content

67%Weight 40%Scale 1-5

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

The body delivers excellent, fully executable docx-js guidance with dense library-specific rules that Claude would not reliably know, and both create and read workflows are clear. It is weakened by ~50 lines of generic template boilerplate, a monolithic 215-line inline example, and — most significantly — zero navigation to the bundled office scripts (pack/unpack/validate/comment), leaving the bundle's most powerful capabilities undiscoverable.

Suggestions

Reference the bundled scripts explicitly so they are discoverable, e.g., "To edit an existing .docx: unpack with scripts/office/unpack.py, edit the XML, validate with scripts/office/validate.py, repack with scripts/office/pack.py; see scripts/comment.py for adding comments".

Move the ~215-line create-doc.js example into a separate reference file (e.g., examples/create-doc.js) and keep a minimal Quick start snippet inline in SKILL.md.

Cut or condense the generic boilerplate sections (Required Inputs, Output Contract, Failure Handling, User Checkpoints, Input Validation, Quick Validation) into a few skill-specific checks — they add ~50 lines of non-docx guidance Claude already handles by default.

Dimension	Reasoning	Score
Conciseness	The docx-specific guidance is dense and non-obvious ("1440 DXA = 1 inch", "prefer ShadingType.CLEAR to avoid unexpected black backgrounds", exact "Heading1"/"Heading2" style IDs), but roughly 50 lines of generic boilerplate (Required Inputs, Output Contract, Failure Handling, User Checkpoints, Input Validation, Quick Validation) pad the file and the create example runs ~215 lines inline. This sits between anchor 2's "several padded sections" and anchor 3's "mostly efficient", landing at 3 since the majority of content is high-signal.	3 / 5
Actionability	Fully executable, copy-paste-ready guidance: a complete create-doc.js program with install and run commands ("npm i docx", "node create-doc.js") covering the common cases (headings, lists, tables, images, TOC, headers/footers, page breaks), plus "pandoc document.docx -o output.md" for extraction and concrete numeric rules in Implementation Details.	5 / 5
Workflow Clarity	The create and read paths are clearly sequenced (Install → create-doc.js → run; pandoc one-liner for reading) and a Quick Validation checklist plus When Not to Use guards provide most checkpoints. Minor gaps keep it below anchor 5: no concrete post-generation verification step (e.g., confirming output.docx opens/renders) and validation guidance remains generic.	4 / 5
Progressive Disclosure	The scripts/ bundle (office/pack.py, office/unpack.py, office/validate.py, comment.py, accept_changes.py) is never referenced anywhere in the body — Quick Validation only vaguely says "Check that key scripts, templates, or reference file paths this skill depends on exist" — and the ~215-line example is inlined monolithically. This matches anchor 2's "references are buried / content that clearly belongs in separate files is inlined"; it cannot be 3 because no references are present at all, let alone weakly signaled.	2 / 5
	Total	14 / 20 Passed

Description

82%Weight 40%Scale 1-5

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 that clearly and explicitly states both the capability set (create and read .docx, generate reports/letters/templates, extract text) and when to invoke it, with the file extension and application named for distinctiveness. Minor deductions for second-person phrasing ("when you need to") and a few missing natural synonyms.

Dimension	Reasoning	Score
Specificity	"Create and read Microsoft Word (.docx) documents" plus "generate reports/letters/templates as .docx or extract readable text" lists several concrete actions (anchor 4), but "Use this skill when you need to" is second-person voice, incurring the rubric's 1-point specificity penalty. It stays above anchor 2 because the actions are concrete, not generic.	3 / 5
Completeness	Explicitly answers both what ("Create and read Microsoft Word (.docx) documents") and when ("Use this skill when you need to generate reports/letters/templates as .docx or extract readable text from existing .docx files") with concrete trigger phrases, matching the anchor 5 example's structure exactly.	5 / 5
Trigger Term Quality	Covers "Microsoft Word", ".docx", "reports/letters/templates", and "extract readable text" — good natural keyword coverage with the file extension included, though common variants like "Word files", bare "docx", or "convert to Markdown" are missing. Anchor 5 requires comprehensive synonym coverage, which this lacks.	4 / 5
Distinctiveness Conflict Risk	"Microsoft Word (.docx) documents" carves out a clear niche with the explicit extension and named application, so it is unlikely to trigger for PDF, presentation, or generic document skills — matching the anchor 5 pattern of a distinct domain with specific triggers.	5 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Other/word-read-write/SKILL.md
Commit: 686e09d

Reviewed: 12 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.