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

88%

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%

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 with concrete rules, thresholds, and explicit feedback loops in its workflows. Its main weakness is size: a 683-line monolith with no bundle files and an unresolved internal reference, where progressive disclosure and token budget could both improve.

Suggestions

Split the large word/phrase tier tables and the context-profile tolerance matrix into a reference bundle file (e.g. references/AUDIT_TABLES.md) and keep SKILL.md as a concise overview pointing one level deep.

Resolve the dangling 'detector/CATEGORIES.md' reference by either creating that bundle file or removing the citation.

Consolidate overlapping pattern entries (e.g. em-dash, bold, and structure rules that recur across sections) to trim length without losing actionable detail.

Dimension	Reasoning	Score
Conciseness	Content is dense and free of concept-explaining fluff, but at ~683 lines the single-file body is far longer than the token budget a lean instruction skill needs; several enumerated patterns could be consolidated. It is mostly efficient but materially over-length for the context window.	2 / 3
Actionability	Guidance is concrete throughout — replace/with tables, regex patterns, numeric thresholds (e.g. '6+ hashtags,' '1 per 1,000 words'), and explicit before/after fixes giving copy-ready direction.	3 / 3
Workflow Clarity	Each mode is a clearly numbered sequence, edit mode ends with an explicit 'Verify' re-read step, and rewrite mode has a second-pass audit feedback loop (audit -> fix -> re-check), matching the validation-checkpoint anchor.	3 / 3
Progressive Disclosure	No bundle files exist, yet the 683-line body is monolithic; a passing reference to 'detector/CATEGORIES.md' points at no real file, and content that could be split (word tiers, tolerance matrix) is inline rather than one level deep.	2 / 3
	Total	10 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is exemplary: it states concrete capabilities, lists natural trigger phrases users would actually say, and explicitly pairs 'what' with 'when' via a Use-when clause. It is distinguishable from other editing skills and uses third-person voice throughout.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — 'Audit and rewrite,' 'detect-only mode,' 'edit-in-place mode,' 'voice profile,' 'iterate-to-convergence pass' — matching the multiple-specific-actions anchor.	3 / 3
Completeness	Explicitly states what the skill does ('Audit and rewrite content to remove AI writing patterns') and when to use it via an explicit 'Use this skill when...' clause with concrete triggers.	3 / 3
Trigger Term Quality	Quotes verbatim natural phrases a user would say ('remove AI-isms,' 'clean up AI writing,' 'audit writing for AI tells,' 'make this sound less like AI'), giving strong coverage of real trigger terms.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (AI-ism removal) with distinctive, specific triggers unlikely to fire for unrelated writing skills.	3 / 3
	Total	12 / 12 Passed

Validation

75%

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 — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (684 lines); consider splitting into references/ and linking	Warning
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	12 / 16 Passed

Repository: pchalasani/claude-code-tools
Path: plugins/writing/skills/remove-ai-patterns/upstream/SKILL.md
Commit: 45ee0e2

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.