avoid-ai-writing-router

Use when a request combines AI-writing audit, rewrite, file editing, voice preservation, false-positive interpretation, verification, or when the user invokes Avoid AI Writing without naming a mode.

Quality

78%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/avoid-ai-writing-router/SKILL.md

Quality

Content

81%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 orchestration skill: bounded multi-stage workflow with explicit validation gates, repair loops, and stop conditions, supported by real one-level-deep reference files. The main improvements are surfacing the bundled validation script and tightening sections that duplicate reference-file detail.

Suggestions

Add a short section or line referencing scripts/validate_connections.py so the bundled validation tooling is discoverable from SKILL.md.

Include a minimal example handoff envelope to make the envelope contract concrete rather than schema-only.

Compress the 'Review lenses' section to name the four lenses and defer their definitions to references/agency-role-lenses.md.

Dimension	Reasoning	Score
Conciseness	The body is dense and rule-driven ('Route to the narrowest owner', a 7-step bounded sequence, an 11-item 'Do not' list) with no explanation of concepts Claude already knows. Minor instances could be trimmed — the handoff envelope field inventory and the four-lens descriptions partially duplicate reference files — so it does not quite reach anchor 5's 'every token earns its place'.	4 / 5
Actionability	Concrete, executable routing directives ('go to `ai-writing-detector`', 'A verifier FAIL returns to the correct repair owner once only if the shared budget has room') plus a worked example request make this mostly actionable instruction. It falls short of anchor 5 because no sample handoff envelope is shown and key parameters like the 'canonical pass cap' resolve only through references.	4 / 5
Workflow Clarity	The multi-stage sequence is explicit with a numbered 7-step flow, a VERIFY gate ('preservation-verifier checks before/after material'), a bounded REPAIR feedback loop ('returns to the correct repair owner once only'), rerun verification, and explicit stop conditions. Mutation is gated by user authorization and host evidence, matching anchor 5's validation and error-recovery pattern.	5 / 5
Progressive Disclosure	The body is a concise overview that clearly signals four one-level-deep references (handoff-contract.md, skill-graph.json, routing-matrix.md, agency-role-lenses.md), all of which exist with real content. The gap keeping it below anchor 5 is that scripts/validate_connections.py and its test exist in the bundle but are never mentioned or surfaced from the body.	4 / 5
	Total	17 / 20 Passed

Description

66%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.

The description has strong, specific trigger coverage and a clear niche, but it is entirely a 'Use when' clause: it never states what the skill actually does, which caps completeness. Adding a one-clause statement of the router's coordinating role would lift it substantially.

Suggestions

Lead with a 'what' clause, e.g. 'Routes and coordinates the Avoid AI Writing skill family (detection, rewriting, file editing, verification, false-positive review).' before the 'Use when' trigger sentence.

Add natural user phrasings as triggers, such as 'humanize', 'make this sound human', or 'sounds AI-generated', to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	The description lists several concrete activity types — 'AI-writing audit, rewrite, file editing, voice preservation, false-positive interpretation, verification' — matching the anchor for several specific actions with minor gaps. It falls short of anchor 5 because it never states what the router itself does (coordinate/route the skill family), only the trigger activities.	4 / 5
Completeness	The 'when' is explicit and detailed ('Use when a request combines... or when the user invokes Avoid AI Writing without naming a mode'), but the 'what' is only implied by the activity list rather than stated. This sits between anchor 2 (only 'when' present, no 'what') and anchor 4 (both present), so a boundary score of 3 applies.	3 / 5
Trigger Term Quality	Natural trigger terms are present ('audit', 'rewrite', 'file editing', 'verification', 'invokes Avoid AI Writing'), giving good keyword coverage. A few common user phrasings are missing ('humanize', 'sounds AI-generated', 'AI detector'), so it does not reach anchor 5's comprehensive synonym coverage.	4 / 5
Distinctiveness Conflict Risk	The 'Avoid AI Writing' niche and terms like 'voice preservation' and 'false-positive interpretation' are distinctive, giving a mostly distinct profile. Generic words ('rewrite', 'file editing', 'verification') create minor overlap risk with general editing skills, keeping it below anchor 5.	4 / 5
	Total	15 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: conorbronsdon/avoid-ai-writing
Path: skills/avoid-ai-writing-router/SKILL.md
Commit: db82bc8

Reviewed: 1 day 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.