Content
25%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill reads more like a product specification document or medical travel guide than an actionable skill for Claude. It is extremely verbose, explaining concepts Claude already knows (disease symptoms, common medications, basic first aid items) while lacking concrete, executable instructions for how to actually perform the described tasks. The absence of bundle files despite numerous data file references, combined with no real validation workflows for health-sensitive operations, significantly undermines its utility.
Suggestions
Drastically reduce content to ~80-100 lines by removing lists of common medications, disease symptoms, and first aid items that Claude already knows. Focus on the unique decision logic and data structures specific to this skill.
Add executable code or concrete step-by-step procedures for each capability instead of vague descriptions. For example, provide actual JSON schema for the tracker file and specific algorithms for risk assessment scoring.
Split detailed reference content (disease databases, drug interaction tables, vaccine schedules) into separate bundle files and reference them from the main SKILL.md.
Add explicit validation checkpoints for health data operations—e.g., verify required fields before generating emergency cards, validate drug interaction checks before outputting recommendations, and include error handling for missing or incomplete health profiles.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is severely verbose at ~350+ lines, extensively listing information Claude already knows (common drug names, basic first aid items, what dengue fever symptoms are, how QR codes work). Much of the content reads like a medical travel guide rather than actionable instructions for Claude. Lists of OTC medications, first aid supplies, and disease incubation periods are general knowledge that doesn't need to be spelled out. | 2 / 5 |
Actionability | Despite its length, the skill provides very little executable guidance. The 'code' examples are mostly pseudocode-level bash comments like 'Read: data/travel-health-tracker.json' without actual implementation. The usage examples show vague input/output descriptions rather than concrete steps Claude should follow. There are no actual functions, API calls, or executable scripts—just templates and descriptions of what outputs should look like. | 2 / 5 |
Workflow Clarity | While the skill lists 7 numbered capabilities, there is no clear workflow sequence for executing any of them. Steps are described at a high level ('analyze,' 'evaluate,' 'generate') without concrete procedures. There are no validation checkpoints despite dealing with health-sensitive data. The data file operations section mentions validation but provides no actual validation logic or error recovery steps. | 2 / 5 |
Progressive Disclosure | The content is a monolithic wall of text with no references to external files despite clearly needing them. The 350+ lines of disease databases, drug lists, and card templates should be split into separate reference files. References to 'data/travel-health-tracker.json' and 'data-example/travel-health-tracker.json' exist but no bundle files are provided, and the data structure is never fully defined in the skill itself. | 2 / 5 |
Total | 8 / 20 Passed |