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

82%

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

65%

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

The body is highly actionable with comprehensive executable examples, but is verbose with some conceptual repetition and inline content that would be better offloaded to the existing reference files. Batch workflows also lack the validation feedback loops the rubric expects.

Suggestions

Move ONNX deployment, batched inference, and the common-workflows section into references/advanced-usage.md to slim the SKILL.md overview and remove inline content that should be separate.

Remove the ASCII architecture diagram and background facts ('1.1 billion masks from 11 million images') that Claude already knows, and de-duplicate the repeated predict() examples.

Add explicit validation checkpoints for batch/destructive operations — e.g., verify mask counts and predicted_iou/stability_score thresholds before accepting results, with a fix-and-retry loop.

Either create scripts/export_onnx_model.py or remove the dangling reference to it so referenced paths resolve to real bundle files.

Dimension	Reasoning	Score
Conciseness	The body is largely executable code, but includes conceptual padding Claude already knows (e.g., the ASCII architecture diagram of Image/Prompt/Mask encoders, 'Trained on 1.1 billion masks from 11 million images') and repeats the predict() call across near-identical examples; it could be tightened.	2 / 3
Actionability	Dense, copy-paste-ready guidance throughout — pip/wget commands, full SamPredictor/SamAutomaticMaskGenerator/ONNX code blocks — with specific parameter values and output schemas rather than pseudocode.	3 / 3
Workflow Clarity	Quick start is well sequenced (install → checkpoints → load → predict), but batch operations (automatic mask generation, batched inference) lack validation/verification checkpoints and there is no validate→fix→retry feedback loop, capping this dimension at 2 per the batch-operations guideline.	2 / 3
Progressive Disclosure	Two well-signaled one-level references exist (references/advanced-usage.md, references/troubleshooting.md), but the ~500-line body keeps substantial sections (ONNX deployment, batched inference, common workflows) inline that could live in advanced-usage.md, and the referenced scripts/export_onnx_model.py path does not exist in the bundle.	2 / 3
	Total	9 / 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.

The description is concrete, trigger-rich, and clearly answers both what the skill does and when to invoke it, using third-person voice consistent with the good examples. It is concise without padding or over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'segment any object in images using points, boxes, or masks as prompts' and 'automatically generate all object masks' — matching the multiple-specific-actions anchor.	3 / 3
Completeness	Clearly 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 an explicit trigger clause.	3 / 3
Trigger Term Quality	Natural user-facing terms are well covered: 'image segmentation', 'segment any object', 'points, boxes, or masks', and 'masks', which a user would plausibly say when needing this skill.	3 / 3
Distinctiveness Conflict Risk	A clear niche (zero-shot image segmentation) with distinct triggers unlikely to fire for unrelated skills; no overlap risk with general vision skills.	3 / 3
	Total	12 / 12 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 (502 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: synthetic-sciences/openscience
Path: backend/cli/skills/llm-tools/segment-anything/SKILL.md
Commit: 52845c3

Reviewed: about 13 hours 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.