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

83%

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

67%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 exceptionally actionable and well-structured with clear mode workflows and validation checkpoints, but it is a monolithic 792-line file with no reference bundle, inverting the progressive-disclosure model where SKILL.md should be an overview pointing to detailed materials.

Suggestions

Move the Tier 1/2/3 word-replacement tables and the long pattern-catalog sections into reference files (e.g. references/word-tiers.md, references/patterns.md) and link to them from a concise overview in SKILL.md.

Trim recurring non-guidance padding — 'Adapted from <repo> PNN' attributions, 'This was implemented as a detector and reverted' meta-notes, and verbose 'Distinct from...' cross-references — or relocate them to a references file.

Extract the context-profile tolerance matrix and voice-profile definitions into references/profiles.md, keeping only the auto-detection cues and default in SKILL.md.

Dimension	Reasoning	Score
Conciseness	The body is mostly substantive reference material, but at ~790 lines it carries noticeable explanatory padding — meta-commentary ('This was implemented as a detector and reverted'), 'Adapted from blader/humanizer PXX' attributions, and repeated 'Distinct from...' cross-reference paragraphs on nearly every entry — that could be tightened without losing the guidance.	3 / 5
Actionability	Highly actionable throughout: concrete Replace\|With word tables, executable validator command ('node detector/validate.js <original> <rewritten>'), explicit regex patterns, and precise per-mode output formats that are copy-paste ready.	5 / 5
Workflow Clarity	Each mode has a numbered job sequence with explicit validation checkpoints — edit mode 're-read the file and confirm the flagged patterns are resolved' and the second-pass audit is a fix-and-retry feedback loop — but the cross-cutting triage flow (P0/P1/P2 severity into the audit) is implied rather than fully sequenced.	4 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ are absent) and the entire 792-line pattern catalog — word tables, tier rules, tolerance matrix — is inlined in SKILL.md when it clearly belongs in separate reference files; section headers provide structure, but the content is not split or navigable across files.	2 / 5
	Total	14 / 20 Passed

Description

100%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.

This is an exemplary description: third-person voice, concrete actions, explicit 'Use when...' trigger guidance with multiple natural synonyms, and a well-scoped niche. It answers what and when with no fluff or over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Audit and rewrite content to remove AI writing patterns', 'detect-only mode', 'edit-in-place mode for files', 'voice profile', 'iterate-to-convergence pass' — giving comprehensive coverage of what the skill does.	5 / 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 trigger phrases, matching the top anchor exactly.	5 / 5
Trigger Term Quality	Quotes five natural trigger phrases a user would actually say ('remove AI-isms,' 'clean up AI writing,' 'edit writing for AI patterns,' 'audit writing for AI tells,' 'make this sound less like AI'), covering synonyms comprehensively.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (AI-writing-pattern removal) with distinct, specific triggers that are unlikely to fire for unrelated skills; minimal conflict risk.	5 / 5
	Total	20 / 20 Passed

Validation

68%

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 — 11 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (793 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
relative_links	Relative link issues: 1 missing	Warning

	Total	11 / 16 Passed

Repository: pchalasani/claude-code-tools
Path: plugins/writing/skills/remove-ai-patterns/upstream/SKILL.md
Commit: c35a135

Reviewed: about 14 hours 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.