docx

Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks

1.28x

Quality

70%

Does it follow best practices?

Impact

100%

1.28x

Average score across 5 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./src/ii_agent/agents/skills/builtin/docx/SKILL.md

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

Quality

Content

52%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 is well-structured with a useful decision tree, concrete commands, and a verification-backed redlining workflow, but its cross-file references are broken: it cites ooxml.md, docx-js.md, and ooxml/scripts/*.py that are absent from the bundle while ignoring the actual Document library in scripts/document.py. This undermines both actionability and progressive disclosure.

Suggestions

Fix the script paths and references: point to the actual bundle (scripts/unpack.py, scripts/pack.py, scripts/document.py) instead of the nonexistent ooxml/scripts/ and ooxml.md, or add those files to the bundle.

Inline a minimal Document-library usage example (e.g. get_node + doc.save() + add_comment from scripts/document.py) so the core editing flow is executable without relying on a missing reference file.

De-duplicate the batch-organization guidance (currently described in both step 2 and step 4) and consolidate the repeated "MANDATORY - READ ENTIRE FILE" instructions to tighten token use.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with a decision tree and terse commands, but batch-organization guidance is duplicated (step 2 "Batch organization" and step 4 "Suggested batch groupings") and the "MANDATORY - READ ENTIRE FILE" phrasing repeats three times — padding that could be tightened, matching the score-3 anchor.	3 / 5
Actionability	Concrete commands (pandoc, soffice, pdftoppm) and a detailed BAD/GOOD redlining XML example are present, but the core editing API (get_node, doc.save(), add_comment) is deferred entirely to ooxml.md — which does not exist in the bundle — and the cited ooxml/scripts/*.py paths do not match the actual scripts/ bundle, leaving key execution details missing.	3 / 5
Workflow Clarity	The redlining workflow has a clear numbered sequence with a final verification step (grep checks for applied/unintended changes) and batch-testing guidance, so verification is not missing; it sits just below score 5 because per-batch validation is described only as "test each batch" without a concrete checkpoint command.	4 / 5
Progressive Disclosure	The body points to one-level-deep references (docx-js.md, ooxml.md), but scored against the actual bundle those referenced files do not exist, while the real bundle (scripts/document.py, scripts/utilities.py, scripts/templates/*.xml) is never referenced — navigation from SKILL.md to the detailed material is broken.	2 / 5
	Total	12 / 20 Passed

Description

88%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 explicitly covers both capabilities and trigger conditions with concrete, mostly-natural phrasing. It is held back from full marks only by a slightly generic "professional documents" term and a broad catch-all clause.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "document creation, editing, and analysis", "tracked changes, comments, formatting preservation, and text extraction" — giving comprehensive coverage rather than just 1-2 actions, matching the score-5 anchor.	5 / 5
Completeness	Explicitly answers both what (creation/editing/analysis with tracked changes, comments, formatting, extraction) and when ("When Claude needs to work with professional documents (.docx files) for: (1)... (4)...") with concrete trigger phrases, matching the score-5 anchor closely.	5 / 5
Trigger Term Quality	Includes natural terms users say ("tracked changes", "comments", ".docx files", "creating/editing documents") with the file extension, but misses common synonyms like "Word documents" or "redlining", so it sits just below comprehensive.	4 / 5
Distinctiveness Conflict Risk	The (.docx files) qualifier carves a clear niche with distinct triggers, but "professional documents" and the trailing "or any other document tasks" over-claim broadly, leaving minor overlap risk versus a score-5 niche.	4 / 5
	Total	18 / 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
relative_links	Relative link issues: 3 missing	Warning

	Total	15 / 16 Passed

Repository: Intelligent-Internet/ii-agent
Path: src/ii_agent/agents/skills/builtin/docx/SKILL.md
Commit: 28563d6

Reviewed: 2 days 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.