writing-voice

Guide sentence-level expression while drafting or revising substantial prose so it remains recognizable as the writer: their vocabulary, humor, certainty, emotional emphasis, and allocation of attention. Use for prose that feels generic, corporate, AI-shaped, over-sanitized, or that no longer sounds like me; for tone and ear passes; and for local wording work in UI copy, documentation, or messages. Do not use to decide a document's content, structure, or destination.

Quality

86%

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

Quality

Content

78%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.

A well-organized, intelligent instruction-only skill with concrete punctuation and rehearsal guidance and a real feedback loop. Its main weakness is that the workflow is implicit in prose rather than crisply sequenced, and some conceptual passages could be more directive.

Suggestions

Convert the Draft-Then-Rehear section into an explicit numbered sequence (draft → read at speaking speed → mark generic tissue/flattened rhythm → remove unsupported polish → on deeper mismatch, reread full source and draft) so the workflow and checkpoints are unambiguous.

Tighten the 'Product Promise' and 'Preserve The Grain' openings into direct imperatives, trimming conceptual framing that does not change what Claude does.

Add one short before/after wording example showing a preserved-voice local revision versus a normalized one, to make the actionability concrete for the most common case.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's intelligence without explaining basic concepts, but some conceptual framing ('Voice is not polish applied after meaning and structure...') could be tightened into more direct directives.	4 / 5
Actionability	Concrete directives abound — 'Prefer periods. Use colons for a real explanation, semicolons for closely related independent clauses', 'read it at speaking speed. Mark generic connective tissue' — though as an instruction-only skill it lacks copy-paste examples; guidance is mostly executable with minor gaps.	4 / 5
Workflow Clarity	A clear draft-then-rehear sequence with a feedback loop ('When a reaction reveals a deeper mismatch, reread the complete source and the whole draft') is present, but it is conveyed through section structure rather than explicit numbered steps or checkpoints.	4 / 5
Progressive Disclosure	Under 50 lines with no need for external references and four well-organized, clearly-headed sections (Product Promise, Preserve The Grain, Draft Then Rehear, Match The Destination), satisfying the simple-skill exception for top marks.	5 / 5
	Total	17 / 20 Passed

Description

95%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 precise, well-triggered description that clearly states what the skill does, when to use it with natural user phrases, and where it stops relative to a sibling skill. The only weakness is that the action list is facet-oriented rather than a set of discrete operations, keeping specificity just below the top anchor.

Dimension	Reasoning	Score
Specificity	Names the domain (drafting/revising substantial prose) and several concrete facets it preserves — 'vocabulary, humor, certainty, emotional emphasis, and allocation of attention' — plus local wording work in UI copy, documentation, or messages; minor gaps versus the most comprehensive anchor.	4 / 5
Completeness	Explicitly answers both what ('Guide sentence-level expression...') and when ('Use for prose that feels generic, corporate, AI-shaped...') with concrete triggers, and adds a negative boundary ('Do not use to decide a document's content, structure, or destination').	5 / 5
Trigger Term Quality	Comprehensive natural trigger phrases a user would actually say — 'generic, corporate, AI-shaped, over-sanitized', 'no longer sounds like me', 'tone and ear passes' — with synonyms across prose, UI copy, documentation, and messages.	5 / 5
Distinctiveness Conflict Risk	Clear niche (sentence-level voice) with an explicit boundary separating it from the sibling 'destination' skill that owns content, structure, and ordering, minimizing wrong-skill triggering.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: EpicenterHQ/epicenter
Path: .agents/skills/writing-voice/SKILL.md
Commit: abe10f3

Reviewed: 5 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.