databricks-unstructured-pdf-generation
Generate PDF documents from HTML and upload to Unity Catalog volumes. Use for creating test PDFs, demo documents, reports, or evaluation datasets.
70
62%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./databricks-skills/databricks-unstructured-pdf-generation/SKILL.mdQuality
Discovery
75%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description is functional and well-structured with clear 'what' and 'when' components, and occupies a distinct niche combining PDF generation with Unity Catalog uploads. However, it could benefit from more specific concrete actions and richer trigger terms to improve discoverability when users phrase requests in varied ways.
Suggestions
Add more specific concrete actions, e.g., 'Converts HTML templates to PDF, supports styling and formatting, uploads generated files to Databricks Unity Catalog volumes'.
Expand trigger terms to include natural variations like 'HTML to PDF', 'Databricks volumes', '.pdf files', 'sample data generation', or 'document pipeline'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (PDF generation from HTML, Unity Catalog volumes) and some actions (creating test PDFs, demo documents, reports), but doesn't list specific concrete actions like 'convert HTML templates to PDF', 'upload to specified volume paths', or 'batch generate documents'. | 2 / 3 |
Completeness | Clearly answers both 'what' (generate PDF documents from HTML and upload to Unity Catalog volumes) and 'when' (Use for creating test PDFs, demo documents, reports, or evaluation datasets), with an explicit 'Use for...' clause providing trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes some relevant keywords like 'PDF', 'HTML', 'Unity Catalog volumes', 'test PDFs', 'reports', and 'evaluation datasets', but misses common variations users might say such as 'generate PDF', 'create document', 'HTML to PDF', '.pdf', or 'Databricks volumes'. | 2 / 3 |
Distinctiveness Conflict Risk | The combination of PDF generation from HTML specifically with Unity Catalog volumes upload is a very distinct niche. This is unlikely to conflict with general PDF reading/editing skills or general file upload skills due to the specific pipeline described. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill is highly actionable with concrete, executable examples and a clear tool signature, but suffers significantly from verbosity. Multiple near-identical HTML examples bloat the content without adding meaningful new information. The workflow section lacks error-handling feedback loops for batch operations, and the lengthy content would benefit from splitting templates and patterns into separate reference files.
Suggestions
Reduce the three 'Common Patterns' to one example and a brief note that Claude should adapt HTML/CSS to the document type — Claude already knows how to write HTML for different domains.
Collapse the parallel execution section to show 1-2 calls with a note like 'Call all N in parallel' rather than listing 5 nearly identical invocations.
Move the professional document template and CSS reference (supported/unsupported features) to a separate TEMPLATES.md or REFERENCE.md file, linked from the main skill.
Add explicit error-handling guidance in the workflow: check the `success` field in each response, retry or report failures, and show an example of handling a failed upload.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~200+ lines. The parallel execution section repeats 5 nearly identical tool calls. The three 'Common Patterns' (technical docs, business reports, HR policies) are essentially the same tool call with different HTML content — Claude doesn't need three examples of writing HTML with different CSS. The professional document template is also unnecessarily long. | 1 / 3 |
Actionability | The tool signature, return format, and all examples are concrete and copy-paste ready. Every pattern shows a complete, executable tool call with full HTML content. The quick start example is immediately usable. | 3 / 3 |
Workflow Clarity | The 'Workflow for Multiple Documents' section provides a clear 4-step sequence, but lacks explicit validation/error-handling checkpoints. Step 4 says 'report results' but doesn't show how to check the response's success/error fields or what to do on failure. For batch operations, this missing feedback loop caps the score at 2. | 2 / 3 |
Progressive Disclosure | The content is structured with clear headers and sections, but it's monolithic — the lengthy HTML templates, three common patterns, and the professional document template could all be in a separate reference file. No external file references are used despite the content length warranting them. | 2 / 3 |
Total | 8 / 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
- databricks-solutions/ai-dev-kit
- Commit
02aac8c
- 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.