Name: avoid-ai-writing
Rating: 68 (1 reviews)
Author: conorbronsdon

avoid-ai-writing

Audit and rewrite content to remove AI writing patterns ("AI-isms"). Use this skill when asked to "remove AI-isms," "clean up AI writing," "edit writing for AI patterns," "audit writing for AI tells," or "make this sound less like AI." Supports a detect-only mode, an edit-in-place mode for files, an optional voice profile (casual / professional / technical / warm / blunt), and an iterate-to-convergence pass.

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

The canonical home for this skill is avoid-ai-writing in conorbronsdon/avoid-ai-writing

Quality

Content

73%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-structured, highly actionable skill body with strong mode-specific workflows, explicit validation feedback loops, and executable commands tied to real bundled scripts. The main weaknesses are repetition in long procedural paragraphs and references to an examples/ directory that is not present in the bundle.

Suggestions

Tighten the long prose paragraphs (edit-mode rules, automatic marks pass, --style composition) by extracting the repeated exemption list (quotes, code, tables, attributed text) into one canonical statement and back-referencing it, reducing token cost.

Either add the referenced examples/ directory (with at least examples/technical.json and examples/README.md) or reframe those paths as user-supplied-only, so every signaled reference resolves to a real bundle file.

Condense the --style resolution prose (lines 135-141) — the path/bare-name disambiguation, fallback, and mechanic-vs-AI-rule precedence currently span three dense paragraphs and could be a short decision table.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and free of basic-concept padding (no 'what a PDF is' filler), but several long procedural paragraphs — the edit-mode rules (line 44), the automatic marks pass (line 60), and the --style composition prose (lines 131-141) — restate exemptions and mechanics repeatedly and could be tightened without losing clarity.	3 / 5
Actionability	Provides concrete executable commands (node scripts/normalize-quotes.js ..., node scripts/check-style.js ..., node detector/validate.js) with explicit exit-code semantics and flag mappings; minor gaps because the cited scripts exist but the referenced examples/ configs and detector/ engine are absent from the bundle, leaving some invocations unverifiable.	4 / 5
Workflow Clarity	Each mode has an explicit numbered sequence (audit→rewrite→diff→second-pass for rewrite, audit→assess for detect, read→edit-in-place→verify for edit) with a genuine validation feedback loop: re-read after editing, second-pass audit, and the preservation validator that exits non-zero on protected-region violations — a textbook destructive/batch validation checkpoint.	5 / 5
Progressive Disclosure	Good structure with a clearly signaled one-level-deep reference load (references/patterns.md) gated behind reference-loading markers and a well-organized severity-tier catalog inline; held at 4 rather than 5 because references to examples/<name>.json and examples/README.md point at an examples/ directory that does not exist in the bundle, so one signaled reference target is not actually resolvable.	4 / 5
	Total	16 / 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 strong, third-person description that explicitly states what the skill does and when to use it, anchored by concrete quoted trigger phrases. Voice is correct third person ('Audit', 'Supports'), with rich capability enumeration and low conflict risk.

Dimension	Reasoning	Score
Specificity	Lists several concrete capabilities — 'Audit and rewrite content to remove AI writing patterns', 'detect-only mode', 'edit-in-place mode', 'optional voice profile', 'iterate-to-convergence pass' — with only minor coverage gaps (no mention of the --style house-style axis or file-attribute exemptions).	4 / 5
Completeness	Explicitly answers both what ('Audit and rewrite content to remove AI writing patterns') and when ('Use this skill when asked to...') with concrete quoted trigger phrases, matching the score-5 anchor exactly.	5 / 5
Trigger Term Quality	Comprehensive natural trigger phrases verbatim users would say ('remove AI-isms', 'clean up AI writing', 'audit writing for AI tells', 'make this sound less like AI') plus the core term 'AI-isms'; coverage of synonyms and quoted phrasings is thorough.	5 / 5
Distinctiveness Conflict Risk	The niche is clear and specific — removing AI writing patterns with a named term ('AI-isms') and distinctive trigger phrases — making conflict with other skills minimal; no overlap risk with generic editing skills given the explicit AI-tell framing.	5 / 5
	Total	19 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: conorbronsdon/avoid-ai-writing
Path: plugins/avoid-ai-writing/skills/avoid-ai-writing/SKILL.md
Commit: 0c2a936

Reviewed: 3 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.