axiom-vision

Use when implementing ANY computer vision feature — image analysis, pose detection, person segmentation, subject lifting, text recognition, barcode scanning.

Quality

85%

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

70%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured routing skill that clearly dispatches vision tasks to one-level-deep reference files with a strong decision tree. Its main weakness is redundancy across four sections that repeat the same routing, and the absence of any inlined executable guidance.

Suggestions

Collapse the redundancy between the Quick Reference table, Decision Tree, Critical Patterns, and Example Invocations — they route the same tasks to the same three files and could be merged into one decision aid plus a compact example or two.

Add a brief inline code snippet or concrete API name per major branch (e.g. the VNRecognizeTextRequest fast/accurate switch) so the skill is actionable without always opening a reference file.

Verify the referenced paths (skills/vision-framework.md, vision-ref.md, vision-diag.md, media-intelligence.md) resolve within the bundle; none were present in references/, scripts/, or assets/.

Dimension	Reasoning	Score
Conciseness	Each section is lean and assumes Claude's competence, but the Quick Reference table, Decision Tree, Critical Patterns, and Example Invocations largely restate the same routing to the same three files, so it could be tightened.	2 / 3
Actionability	Routing is concrete and specific — a clear decision tree maps each condition to a specific file — but the body contains no executable code or commands, delegating all implementation to referenced files.	2 / 3
Workflow Clarity	The decision flow is unambiguous and well-sequenced via the dot graph plus a numbered list that maps each task type to one file; no validation checkpoints are needed for a pure routing skill.	3 / 3
Progressive Disclosure	SKILL.md acts as an overview with well-signaled, one-level-deep references (vision-framework.md, vision-ref.md, vision-diag.md, media-intelligence.md) organized in a navigable Quick Reference table.	3 / 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.

A strong, concise description that names concrete capabilities, includes an explicit 'Use when' trigger, and carves out a distinct computer-vision niche. It does not pad with buzzwords or over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'image analysis, pose detection, person segmentation, subject lifting, text recognition, barcode scanning' — matching the level-3 anchor that asks for several specific concrete actions.	3 / 3
Completeness	Explicitly answers both what (the enumerated vision capabilities) and when via the 'Use when implementing ANY computer vision feature' trigger clause, matching the level-3 anchor.	3 / 3
Trigger Term Quality	Uses natural terms users would say ('pose detection', 'person segmentation', 'barcode scanning', 'computer vision feature') with good coverage across the domain, matching the level-3 anchor.	3 / 3
Distinctiveness Conflict Risk	Scopes to 'computer vision' with the Vision framework as a clear niche and distinct triggers, making it unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 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: CharlesWiltgen/Axiom
Path: .claude-plugin/plugins/axiom/skills/axiom-vision/SKILL.md
Commit: ea3be7c

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.