imaging-data-commons

Query and download public cancer imaging data from NCI Imaging Data Commons using idc-index. Use for accessing large-scale radiology (CT, MR, PET) and pathology datasets for AI training or research. No authentication required. Query by metadata, visualize in browser, check licenses.

Quality

86%

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

72%

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

A highly actionable, well-structured body with excellent progressive disclosure and executable examples throughout. Its main weaknesses are length/redundancy around version verification and the absence of explicit validation feedback loops in batch/download workflows.

Suggestions

Consolidate the version-verification guidance into one location (e.g., Installation) and reference it elsewhere, removing the repeated upgrade snippets in Overview and Best Practices to tighten conciseness.

Add an explicit validate→fix→retry feedback loop to the batch download workflow (e.g., verify each batch's file count/integrity before proceeding to the next, and retry failed series).

Consider moving the full inline index/join tables into references/index_tables_guide.md, keeping only the most-used tables inline with a pointer, to reduce SKILL.md length while preserving the strong navigation.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient, concrete, and free of basic-concept filler, but at ~850 lines it repeats version-verification guidance in three places (Overview, Installation, Best Practices) and could be tightened; it is not score 3 because not every token earns its place, and not score 1 because it avoids explaining concepts Claude already knows.	2 / 3
Actionability	Provides fully executable, copy-paste-ready Python, SQL, and bash examples throughout (sql_query, download_from_selection, get_viewer_URL, idc CLI, manifest generation), matching the score-3 anchor of concrete executable guidance with specific examples.	3 / 3
Workflow Clarity	Steps are sequenced (core workflow 1-2-3, explore-then-query, version-check-first) with some checkpoints, but batch and bulk-download operations lack explicit validate→fix→retry feedback loops; per the rubric, missing validation in batch/destructive operations caps this at 2 rather than 3.	2 / 3
Progressive Disclosure	A Quick Navigation section lists inline core sections plus a 'Reference Guides' table with a 'When to Load' column, and all 10 references/ files exist and are one level deep with clear 'See references/X.md for...' signals, matching the score-3 anchor of well-signaled one-level-deep references with easy navigation.	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, third-person description that clearly states capabilities and provides an explicit 'Use for' trigger with good natural keyword coverage and a distinct, low-conflict niche. No significant weaknesses; voice and trigger guidance both meet the highest anchors.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Query and download', 'Query by metadata, visualize in browser, check licenses' — tied to a specific domain (radiology CT/MR/PET and pathology datasets), matching the score-3 anchor of several specific concrete actions.	3 / 3
Completeness	Explicitly answers both what ('Query and download public cancer imaging data from NCI Imaging Data Commons using idc-index') and when via an explicit 'Use for...' trigger clause, satisfying the score-3 anchor; it is not score 2 because the when is stated explicitly rather than implied.	3 / 3
Trigger Term Quality	Covers natural terms users would say — 'cancer imaging data', 'NCI Imaging Data Commons', 'radiology (CT, MR, PET)', 'pathology datasets', 'AI training or research', 'check licenses' — giving broad natural-language coverage rather than jargon.	3 / 3
Distinctiveness Conflict Risk	The 'NCI Imaging Data Commons' / 'idc-index' / 'cancer imaging data' niche is highly specific with distinct triggers, making conflict with other skills unlikely; it is well above the score-2 'could overlap' anchor.	3 / 3
	Total	12 / 12 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
skill_md_line_count	SKILL.md is long (858 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: K-Dense-AI/scientific-agent-skills
Path: skills/imaging-data-commons/SKILL.md
Commit: 757b63b

Reviewed: about 11 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.