dicom-anonymizer
Batch anonymize DICOM medical images by removing patient sensitive information (name, ID, birth date) while preserving image data for research use. Trigger when users need to de-identify medical imaging data, prepare DICOM files for research sharing, or remove PHI from radiology/scanned images.
90
Quality
86%
Does it follow best practices?
Impact
100%
1.51xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
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.
This is an excellent skill description that clearly defines a specialized capability for DICOM medical image anonymization. It provides specific actions, comprehensive trigger terms that domain users would naturally use, and explicitly states both what the skill does and when to use it. The medical imaging focus makes it highly distinctive from other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'batch anonymize', 'removing patient sensitive information (name, ID, birth date)', 'preserving image data for research use'. The parenthetical examples add helpful specificity. | 3 / 3 |
Completeness | Clearly answers both what ('Batch anonymize DICOM medical images by removing patient sensitive information...') AND when ('Trigger when users need to de-identify medical imaging data, prepare DICOM files for research sharing, or remove PHI...'). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: 'DICOM', 'medical images', 'de-identify', 'medical imaging data', 'research sharing', 'PHI', 'radiology', 'scanned images'. These are terms domain users would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche targeting DICOM files specifically, medical imaging anonymization, and PHI removal. Very unlikely to conflict with general document or image processing skills due to specialized medical/research focus. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
72%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides strong actionable guidance with executable code examples and comprehensive parameter/tag documentation. However, it suffers from boilerplate bloat (security checklists, lifecycle status, evaluation criteria) that don't add value for Claude, and the workflow lacks explicit validation checkpoints critical for PHI-handling batch operations.
Suggestions
Remove boilerplate sections (Security Checklist, Risk Assessment, Evaluation Criteria, Lifecycle Status) that don't provide actionable guidance for Claude
Add an explicit validation workflow with checkpoints: e.g., '1. Run on single file first, 2. Verify audit log, 3. Spot-check output DICOM, 4. Only then run batch'
Remove the Overview paragraph explaining what DICOM is - Claude already knows this
Integrate the 'Always perform manual QA' warning into a concrete workflow step rather than a standalone warning
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill contains significant padding including boilerplate sections (Security Checklist, Risk Assessment, Evaluation Criteria, Lifecycle Status) that add little value. The Overview section explains what DICOM is, which Claude already knows. However, the core technical content (tag tables, parameters) is reasonably efficient. | 2 / 3 |
Actionability | Provides fully executable CLI commands and Python API examples that are copy-paste ready. The parameter table is comprehensive with clear types and defaults. The tag tables specify exact DICOM tag codes and actions taken. | 3 / 3 |
Workflow Clarity | The Technical Architecture section lists steps but lacks explicit validation checkpoints. For a tool handling PHI with destructive batch operations, there's no clear 'validate -> fix -> retry' feedback loop. The warnings mention manual QA but don't integrate it into a workflow sequence. | 2 / 3 |
Progressive Disclosure | Good structure with clear sections progressing from usage to parameters to technical details. References to external files (requirements.txt, phi_tags.json, PDF guides) are one level deep and clearly signaled. Content is appropriately split between quick-start usage and detailed reference tables. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- aipoch/medical-research-skills
- Commit
ca9aaa4
- Reviewed
Table of Contents
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.