imaging-data-commons

Query and download public cancer imaging data from NCI Imaging Data Commons. Invoke for any question about IDC collections, cancer imaging datasets, DICOM data access, radiology (CT, MR, PET) or pathology AI training sets, metadata queries, visualization, or license checks — even when the user doesn't explicitly mention "IDC". No authentication required.

Quality

96%

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

93%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 well-engineered skill body: concise, fully actionable, and excellently structured with real on-demand reference guides. The only gap is that batch/destructive workflows rely on advisory tips rather than explicit validate-and-retry feedback loops.

Suggestions

Add an explicit validate->fix->retry loop for batch downloads (e.g., after download_from_selection, verify file count/openability with pydicom.dcmread and re-run for any missing or unreadable series) to satisfy the feedback-loop guidance for destructive/batch operations.

For mixed-license cohorts, add a checkpoint that fails the workflow when the most restrictive governing license is detected (e.g., explicit halt on CC BY-NC for commercial use) rather than only describing the rule.

Make the IDC data version verification a mandatory first step with a clear halt-and-escalate instruction when the local index is a whole release behind, rather than an advisory best-practice bullet.

Dimension	Reasoning	Score
Conciseness	Lean and efficient throughout: it assumes Claude's competence (no "what a DICOM is" filler), front-loads a routing gate, and defers detail to on-demand reference guides; every token earns its place.	5 / 5
Actionability	Fully executable copy-paste code and curl commands cover the common cases (sql_query, download_from_selection, download_dicom_series, get_viewer_URL, licenses/citations, REST endpoints), with a concrete table disambiguating the two download methods' argument order.	5 / 5
Workflow Clarity	The routing gate is a well-sequenced decision tree and the download workflow is explicit, but batch/destructive operations (bulk download, mixed-license cohorts) lack an explicit validate->fix->retry feedback loop rather than just advisory batching/resume tips, so it sits just below the top anchor with the validation cap in mind.	4 / 5
Progressive Disclosure	Clear overview with a well-signaled, one-level-deep reference table; all 13 referenced guides and scripts/check_version.py exist as real bundle files, and content is appropriately split between inline essentials and on-demand guides.	5 / 5
	Total	19 / 20 Passed

Description

100%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.

An exemplary description: concrete actions, comprehensive natural trigger terms, explicit what-and-when guidance, and a clearly scoped niche. It even anticipates the most likely miss-trigger (users not saying 'IDC') and pre-empts it.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ("Query and download", "metadata queries, visualization, or license checks") with comprehensive coverage of the skill's capabilities, matching the top anchor.	5 / 5
Completeness	Explicitly answers both what ("Query and download public cancer imaging data") and when ("Invoke for any question about..."), with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Comprehensive natural-term coverage including synonyms and abbreviations ("IDC collections", "cancer imaging datasets", "DICOM data access", "radiology (CT, MR, PET)", "pathology AI training sets", "license checks"), matching the top anchor.	5 / 5
Distinctiveness Conflict Risk	Clear niche (NCI Imaging Data Commons cancer imaging) with distinct triggers and even a guard against the most common misfire ("even when the user doesn't explicitly mention 'IDC'"), minimal conflict risk.	5 / 5
	Total	20 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: K-Dense-AI/scientific-agent-skills
Path: skills/imaging-data-commons/SKILL.md
Commit: f6fcafe

Reviewed: ~23 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.