Name: computer-vision-expert
Rating: 43.2 (1 reviews)
Author: administrakt0r

computer-vision-expert

SOTA Computer Vision Expert (2026). Specialized in YOLO26, Segment Anything 3 (SAM 3), Vision Language Models, and real-time spatial analysis.

Quality

43%

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 ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/computer-vision-expert/SKILL.md

The canonical home for this skill is computer-vision-expert in administrakt0r/AI-Agents-Safe-Coding-Skills

Quality

Content

47%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 a well-structured, token-efficient overview of the CV niche, but it stays entirely at the conceptual level: no executable code or commands and no sequenced workflows with validation. It works as a map but not as an actionable guide.

Suggestions

Add concrete, copy-paste-ready examples for the common cases — e.g. a YOLO26 inference + ONNX/TensorRT export snippet and a SAM 3 text-to-mask call.

Convert the Patterns into numbered workflows with explicit validation steps (e.g. verify detection confidence before mask refinement, validate export output on target hardware).

Move detailed per-model capability notes into reference files (e.g. references/yolo26.md, references/sam3.md) and keep SKILL.md as a concise overview with one-level-deep links.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean bullet points that assume Claude's familiarity with the models, with only minor trimmable puffery ("Mastery of", "State-of-the-art", "2x accuracy over SAM 2") and light redundancy across Capabilities, Patterns, and Anti-Patterns.	4 / 5
Actionability	The skill gives only high-level hints ("Combine YOLO26 for fast 'candidate proposal' and SAM 3 for 'precise mask refinement'") with no code, commands, configs, or executable examples anywhere in the body.	2 / 5
Workflow Clarity	The Patterns section offers rough sequences (candidate proposal then mask refinement; depth maps with homographies) but presents them as prose with poorly defined steps and no validation checkpoints or feedback loops.	2 / 5
Progressive Disclosure	Content is well organized into single-level sections (Purpose, When to Use, Capabilities, Patterns, Anti-Patterns, Sharp Edges) with no nested file references; it is slightly over the 50-line simple-skill threshold and could offload some capability detail to reference files.	4 / 5
	Total	12 / 20 Passed

Description

40%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 reads as a role title rather than a functional trigger: it names specialized technologies but states no concrete actions and provides no "Use when" guidance. It is distinctive within its niche yet weak on completeness and trigger phrasing.

Suggestions

Rewrite the description using third-person action verbs, e.g. "Designs and optimizes computer vision pipelines: real-time detection with YOLO26, promptable segmentation with SAM 3, and visual reasoning with VLMs."

Add an explicit trigger clause such as "Use when building real-time object detection, text-guided segmentation, depth estimation, or edge-deployed vision systems."

Include common user-facing synonyms (object detection, image segmentation, depth estimation, edge deployment) alongside the model names to improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain and specific technologies ("YOLO26, Segment Anything 3 (SAM 3), Vision Language Models, and real-time spatial analysis") but uses the persona phrase "Specialized in" rather than concrete action verbs like the good examples ("Extract", "Analyze", "Generate"); no actions are actually listed.	2 / 5
Completeness	There is no "Use when..." clause or equivalent trigger guidance (which caps completeness at 3), and the "what" is a vague role statement ("SOTA Computer Vision Expert... Specialized in...") rather than a functional action description — fitting the vague-what-plus-no-when anchor.	2 / 5
Trigger Term Quality	"Computer Vision" and "real-time spatial analysis" are natural terms a user might say, but the description leans heavily on jargon (YOLO26, SAM 3, VLMs) and omits common synonyms such as object detection, image segmentation, or face recognition.	3 / 5
Distinctiveness Conflict Risk	The computer-vision niche anchored to specific SOTA models (YOLO26, SAM 3, VLMs) is mostly distinct with only minor overlap risk against closely related AI/robotics skills.	4 / 5
	Total	11 / 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: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills-claude/skills/computer-vision-expert/SKILL.md
Commit: 57c135b

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.