pubmed-database
Direct REST API access to PubMed. Advanced Boolean/MeSH queries, E-utilities API, batch processing, citation management. For Python workflows, prefer biopython (Bio.Entrez). Use this for direct HTTP/REST work or custom API implementations.
80
78%
Does it follow best practices?
Impact
72%
1.16xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./scientific-skills/pubmed-database/SKILL.mdQuality
Discovery
85%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 a strong skill description that clearly defines its technical scope (PubMed REST API) and provides explicit guidance on when to use it versus alternatives (biopython). The main weakness is that trigger terms lean heavily technical, which may miss users who describe their needs in more natural language like 'search medical papers' or 'find research articles'.
Suggestions
Add natural language trigger terms that users might say, such as 'medical literature search', 'research papers', 'NCBI database', or 'scientific articles'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'Advanced Boolean/MeSH queries, E-utilities API, batch processing, citation management' and specifies 'Direct REST API access to PubMed' - these are concrete, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers what (REST API access, Boolean/MeSH queries, batch processing, citation management) AND when ('For Python workflows, prefer biopython', 'Use this for direct HTTP/REST work or custom API implementations') with explicit guidance on when to use vs alternatives. | 3 / 3 |
Trigger Term Quality | Includes relevant technical terms like 'PubMed', 'REST API', 'MeSH queries', 'E-utilities', 'HTTP/REST', but these are more technical jargon than natural user language. Missing common variations like 'medical literature', 'research papers', 'NCBI'. | 2 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with clear niche: specifically PubMed REST API access, explicitly differentiates from biopython/Bio.Entrez for Python workflows. The 'direct HTTP/REST work or custom API implementations' clause creates clear boundaries. | 3 / 3 |
Total | 11 / 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 is a comprehensive and well-structured skill with strong actionability through executable code examples and excellent progressive disclosure via clearly organized reference files. The main weaknesses are verbosity in introductory sections that explain concepts Claude already knows, and workflows that lack explicit validation checkpoints for error-prone operations like batch API processing.
Suggestions
Remove or significantly condense the 'Overview' and 'When to Use This Skill' sections - Claude knows what PubMed is and can infer appropriate use cases
Add explicit validation steps to API workflows, e.g., 'Verify response status code before parsing' and 'Check idlist is non-empty before proceeding to EFetch'
Add error handling examples showing how to detect and recover from rate limiting or malformed responses in the Python code
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill contains some unnecessary explanations (e.g., 'PubMed is the U.S. National Library of Medicine's comprehensive database...') and verbose 'When to Use This Skill' section that Claude doesn't need. However, the core technical content is reasonably efficient. | 2 / 3 |
Actionability | Provides fully executable Python code examples for API access, concrete query syntax with real examples, and specific field tags and parameters. The code is copy-paste ready and includes practical details like API keys and rate limits. | 3 / 3 |
Workflow Clarity | Workflows are listed with clear steps but lack explicit validation checkpoints. For API workflows involving batch operations, there's no feedback loop for error recovery or validation steps between operations. The workflows read more like checklists than guided processes with verification. | 2 / 3 |
Progressive Disclosure | Excellent structure with clear overview in SKILL.md and well-signaled references to three specific reference files (api_reference.md, search_syntax.md, common_queries.md). Each reference is described with specific 'consult when' guidance and grep patterns for discovery. | 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 |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
Total | 10 / 11 Passed | |
- Repository
- K-Dense-AI/claude-scientific-skills
- Commit
71add64
- 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.