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

84%

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

76%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 dense, highly actionable reference with executable examples and sensible external references. The main gap is missing validation/feedback checkpoints for batch and destructive operations, which caps workflow clarity.

Suggestions

Add explicit validation/verification steps (e.g., check predicted_iou/stability_score thresholds before accepting masks; assert mask shape/non-empty) to batch and automatic-mask-generation workflows.

Introduce a feedback loop for ONNX export (validate the exported model produces matching outputs vs. the PyTorch model) before deployment.

Move some of the deeper code examples (e.g., ONNX inference plumbing, batched-inference loops) into references/advanced-usage.md to slim the SKILL.md body and strengthen progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Mostly lean and code-forward, assuming Claude's competence; minor background asides (e.g., 'Trained on 1.1 billion masks from 11 million images') could be trimmed but do not pad heavily.	4 / 5
Actionability	Copy-paste-ready, executable code throughout — install, checkpoint download, SamPredictor, HuggingFace transformers, all prompt types, and ONNX export/inference — covering the common cases completely.	5 / 5
Workflow Clarity	Sections are logically sequenced (install → checkpoints → usage → advanced), but batch operations (automatic mask generation, batched inference) lack explicit validation/verification checkpoints, which per the rubric caps this dimension at 3.	3 / 5
Progressive Disclosure	Core content lives in the body with well-signaled one-level-deep references to real files (references/advanced-usage.md, references/troubleshooting.md); good structure, though the body itself is long and could externalize more detail to reach 5.	4 / 5
	Total	16 / 20 Passed

Description

92%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.

A strong, specific description that clearly states capabilities and when to invoke the skill. Trigger phrasing is natural but could add synonyms or extension hints for completeness.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks' — covering the main SAM capabilities comprehensively.	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...'), with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural phrasings like 'segment any object in images', 'points, boxes, or masks as prompts', and 'automatically generate all object masks', but lacks synonyms or file-extension variants that would push it to a 5.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (zero-shot image segmentation via SAM) with distinct triggers and minimal overlap risk with other skills.	5 / 5
	Total	19 / 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 (510 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	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	11 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/llm-tools/segment-anything/SKILL.md
Commit: 543b17f

Reviewed: 5 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.