devtu-docs-quality

TOP PRIORITY skill — find and immediately fix or remove every piece of wrong, outdated, or redundant information in ToolUniverse docs. Wrong code, broken links, incorrect counts, and overlapping instructions must be fixed or removed — never left in place. Runs five phases: (D) static method scan, (C) live code execution, (A) automated validation, (B) ToolUniverse audit, (E) less-is-more simplification. Core philosophy: each concept appears exactly once; remove don't add; no emojis; single setup entry point. Use when reviewing docs, before releases, after API changes, or when asked to audit, fix, or simplify documentation.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%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 highly actionable and the five-phase workflow is exceptionally well sequenced with exit-code gates and a fix-and-retry checklist. Its weaknesses are self-inflicted: it repeats its own rules in multiple sections despite mandating 'each concept appears exactly once', and it references bundle files that are not present while inlining two large scripts that belong in scripts/.

Suggestions

Move the Phase D and Phase E scanner scripts into scripts/ files (mirroring test_doc_code_blocks.py) and invoke them by command, removing ~125 lines of inline Python from SKILL.md.

Actually include API_REFERENCE.md, DOCS_STRUCTURE.md, and scripts/test_doc_code_blocks.py in the skill bundle, or remove the 'Reference Files' section — the current references are dead paths.

Consolidate the duplicated rules (no emojis, '1000+ tools', single setup entry point) into the 'Apple-Style Simplification Rules' section and reference them from the checklist instead of restating them.

Dimension	Reasoning	Score
Conciseness	The scanners, fix tables, and checklists are dense and project-specific, but the body violates its own 'each concept appears exactly once' rule: the no-emoji rule, the '1000+ tools' rule, and the setup entry point are each stated in 2-3 places, and the philosophy is duplicated across 'Two Equal Priorities' and 'Less Is More'.	3 / 5
Actionability	Fully executable throughout: complete heredoc Python scanners, exact commands (e.g., the rg tool-count sweep), a wrong-to-correct method fix table, and a runtime-failure table with causes and fixes.	5 / 5
Workflow Clarity	Phases are explicitly ordered ('D first, then E, then C/A/B') with a summary table, exit-code gates on every scan, and a validation checklist with an explicit feedback loop ('stop, fix it, re-run the relevant phase, confirm it passes').	5 / 5
Progressive Disclosure	A clearly signaled 'Reference Files' section exists, but API_REFERENCE.md, DOCS_STRUCTURE.md, and scripts/test_doc_code_blocks.py are absent from the bundle, and ~125 lines of scanner Python are inlined for Phases D/E while Phase C is (inconsistently) invoked as a script file.	3 / 5
	Total	16 / 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: it is written in third person, states concrete actions with a comprehensive failure taxonomy and phase list, and closes with an explicit 'Use when...' clause covering reviews, releases, API changes, and audit/fix/simplify requests. The only weaknesses are a few missing natural synonyms and slightly generic trigger phrases that could overlap non-ToolUniverse doc tasks.

Dimension	Reasoning	Score
Specificity	Names concrete failure classes ('Wrong code, broken links, incorrect counts, and overlapping instructions') and enumerates all five phases (D/C/A/B/E), giving comprehensive coverage of what the skill does.	5 / 5
Completeness	Explicitly answers both what ('find and immediately fix or remove every piece of wrong, outdated, or redundant information in ToolUniverse docs') and when (a full 'Use when...' clause with concrete trigger situations).	5 / 5
Trigger Term Quality	'Use when reviewing docs, before releases, after API changes, or when asked to audit, fix, or simplify documentation' covers natural phrases well, but misses common variations like 'docs cleanup', 'stale docs', or 'proofread'.	4 / 5
Distinctiveness Conflict Risk	The 'ToolUniverse docs' scope carves a clear niche, but generic triggers ('reviewing docs', 'simplify documentation') create minor overlap risk with general documentation-review skills.	4 / 5
	Total	18 / 20 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
relative_links	Relative link issues: 2 missing	Warning
referenced_paths_exist	Referenced path issues: 5 missing	Warning

	Total	14 / 16 Passed

Repository: mims-harvard/ToolUniverse
Path: skills/devtu-docs-quality/SKILL.md
Commit: 164aaf1

Reviewed: 6 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.