extract-from-pdfs
This skill should be used when extracting structured data from scientific PDFs for systematic reviews, meta-analyses, or database creation. Use when working with collections of research papers that need to be converted into analyzable datasets with validation metrics.
72
87%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
89%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 solid description with strong trigger terms and clear 'when to use' guidance targeting a well-defined niche in scientific literature processing. Its main weakness is that the 'what' portion could be more specific about the concrete actions performed (e.g., extracting study characteristics, parsing statistical results, coding effect sizes). The domain-specific terminology provides excellent distinctiveness.
Suggestions
Add more specific concrete actions to improve specificity, e.g., 'Extracts study characteristics, statistical results, sample sizes, and methodology details from scientific PDFs' rather than the general 'extracting structured data'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain (scientific PDFs, systematic reviews) and mentions some actions (extracting structured data, converting to analyzable datasets), but doesn't list multiple specific concrete actions like parsing tables, extracting methodology sections, coding effect sizes, or identifying study characteristics. | 2 / 3 |
Completeness | The description explicitly answers both 'what' (extracting structured data from scientific PDFs, converting research papers into analyzable datasets with validation metrics) and 'when' ('Use when working with collections of research papers that need to be converted into analyzable datasets'). It has clear 'Use when' guidance. | 3 / 3 |
Trigger Term Quality | Good coverage of natural terms users would say: 'scientific PDFs', 'systematic reviews', 'meta-analyses', 'database creation', 'research papers', 'analyzable datasets', 'structured data', 'validation metrics'. These are terms researchers would naturally use when requesting this kind of work. | 3 / 3 |
Distinctiveness Conflict Risk | The description carves out a clear niche: scientific PDF extraction specifically for systematic reviews and meta-analyses. This is distinct from general PDF processing skills or generic data extraction skills, with specific triggers like 'systematic reviews' and 'meta-analyses' that are unlikely to conflict with other skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, highly actionable skill that provides a clear multi-step pipeline with executable commands and good progressive disclosure to reference materials. Its main weakness is moderate verbosity—some sections duplicate information (Available Scripts repeats the workflow), and explanations of basic statistical concepts (precision, recall, F1) are unnecessary for Claude. Trimming redundant sections and removing known-concept explanations would improve token efficiency.
Suggestions
Remove the 'Available Scripts' section entirely since every script is already shown with full usage in the 'Workflow Execution' section—this is pure duplication.
Remove explanations of precision, recall, and F1 score definitions in the Quality Assurance section; Claude already knows these concepts. Keep only the actionable guidance on sample sizes and how to use metrics.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some sections that could be tightened—e.g., the 'When to Use This Skill' section, the 'Available Scripts' section that duplicates information already shown in the workflow, and the 'Quality Assurance' section which explains basic concepts like precision/recall that Claude already knows. The cost optimization section adds value but is somewhat verbose. | 2 / 3 |
Actionability | The skill provides fully executable, copy-paste-ready command-line invocations for every pipeline step with specific flags and arguments. Schema customization steps are concrete with file paths and field names. The guidance is specific and directly usable. | 3 / 3 |
Workflow Clarity | The 6-step pipeline (plus 3 validation steps) is clearly sequenced with explicit commands at each stage. The workflow includes validation checkpoints (Step 4 repair, Step 5 API validation, Steps 7-9 quality metrics), and the iterative improvement section provides a clear feedback loop for error recovery and quality improvement. | 3 / 3 |
Progressive Disclosure | The skill provides a clear overview with well-signaled one-level-deep references to setup_guide.md, workflow_guide.md, validation_guide.md, and api_reference.md. Templates and examples are referenced by path. The main content stays at the right level of detail while pointing to deeper documentation. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- brunoasm/my_claude_skills
- Commit
2a19b68
- 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.