segment-anything-model

Foundation model for image segmentation with zero-shot transfer. Use when you need to segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks in an image.

Quality

75%

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/multimodal/segment-anything/SKILL.md

Quality

Content

61%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 skill body is highly actionable with comprehensive executable code and well-organized references to real bundle files. It would improve by trimming restated concepts and moving more reference-style sections into the existing reference files.

Suggestions

Trim concepts Claude already knows (e.g. parenthetical explanations and the restated common-issues table) to improve conciseness.

Move API-reference-style sections (ONNX deployment, batched inference, COCO RLE format, output data structure) into references/advanced-usage.md, keeping only a quick example inline.

Add a short explicit validation/checkpoint step for the ONNX export workflow (e.g. verifying the exported model runs) to strengthen workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with extensive code, but includes unnecessary restating of concepts Claude already knows (e.g. "Set image (computes embeddings once)", restated common-issues table) and could be tightened.	3 / 5
Actionability	Provides executable, copy-paste-ready code across the main cases (point/box/combined prompts, automatic generation, ONNX, batched inference) with concrete checkpoint URLs; minor gaps like undefined cv2/image variables in a few snippets.	4 / 5
Workflow Clarity	Topical sequencing (install → checkpoints → usage → advanced) is clear, but it reads as a reference catalog of snippets rather than a sequenced workflow, with no explicit validation checkpoints or feedback loops.	3 / 5
Progressive Disclosure	Good structure with bulk content correctly split into two real referenced files (advanced-usage.md, troubleshooting.md) that are clearly signaled, though a fair amount of API-reference-style content (ONNX, batched inference, output format, COCO RLE) is inlined in SKILL.md.	4 / 5
	Total	14 / 20 Passed

Description

88%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 is specific, complete, and clearly distinguishes the skill's niche with explicit what/when guidance. It could add a few more natural synonyms (semantic/instance segmentation) to round out trigger-term coverage.

Suggestions

Add common synonyms like "semantic segmentation" or "instance segmentation" to broaden trigger-term coverage.

Optionally mention image file types users might name (e.g. .jpg/.png) to strengthen natural triggers.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — interactive segmentation via "points, boxes, or masks as prompts" and "automatically generate all object masks" — giving comprehensive coverage of SAM's capabilities.	5 / 5
Completeness	Explicitly answers both what ("Foundation model for image segmentation with zero-shot transfer") and when ("Use when you need to segment any object... or automatically generate all object masks") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural terms like "image segmentation", "segment any object in images", and the prompt types, but omits common synonyms (semantic/instance segmentation) and file-type triggers, so a few natural terms are missing.	4 / 5
Distinctiveness Conflict Risk	The zero-shot promptable-segmentation niche is distinct from generic CV skills, with only minor overlap risk against closely related skills like SAM 2 or detection pipelines.	4 / 5
	Total	18 / 20 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 (501 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	12 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/multimodal/segment-anything/SKILL.md
Commit: 4b35756

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.