markitdown
Convert files and office documents to Markdown. Supports PDF, DOCX, PPTX, XLSX, images (with OCR), audio (with transcription), HTML, CSV, JSON, XML, ZIP, YouTube URLs, EPubs and more.
74
66%
Does it follow best practices?
Impact
85%
3.40xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./scientific-skills/markitdown/SKILL.mdQuality
Discovery
82%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 description with excellent specificity and trigger term coverage across a wide range of file formats. Its main weakness is the absence of an explicit 'Use when...' clause, which would help Claude know precisely when to select this skill. The description clearly communicates what the skill does but relies on implicit triggers rather than explicit guidance.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user wants to convert a file to Markdown, extract text from documents, transcribe audio, or process any supported file format into readable Markdown.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and formats: 'Convert files and office documents to Markdown' with explicit support for PDF, DOCX, PPTX, XLSX, images with OCR, audio with transcription, HTML, CSV, JSON, XML, ZIP, YouTube URLs, EPubs. | 3 / 3 |
Completeness | Clearly answers 'what does this do' (convert files to Markdown with support for many formats), but lacks an explicit 'Use when...' clause or equivalent trigger guidance. The when is only implied by the list of supported formats. | 2 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: PDF, DOCX, PPTX, XLSX, images, OCR, audio, transcription, HTML, CSV, JSON, XML, ZIP, YouTube, EPub, Markdown, convert. These are all terms users would naturally use when needing file conversion. | 3 / 3 |
Distinctiveness Conflict Risk | The specific niche of converting diverse file formats to Markdown is quite distinct. The combination of 'convert to Markdown' with the extensive format list makes it unlikely to conflict with other skills like general PDF processing or audio transcription tools. | 3 / 3 |
Total | 11 / 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 provides excellent actionable code examples covering a wide range of MarkItDown use cases, but is severely bloated with redundant examples, an irrelevant 'Scientific Schematics' section that appears injected from another skill, and marketing-style explanations. The content would be far more effective at 1/3 its current length with use cases moved to a separate examples file.
Suggestions
Remove the entire 'Visual Enhancement with Scientific Schematics' section - it's unrelated to file-to-markdown conversion and appears to be cross-contamination from another skill.
Consolidate the 6 'Common Use Cases' into 1-2 representative examples in the main file, moving the rest to a separate examples.md reference file.
Remove explanatory text Claude already knows (e.g., 'Markdown is token-efficient', 'Key Benefits' list, 'Performance Considerations' section with obvious statements like 'Large PDFs may take time').
Add a validation step to batch processing workflows - e.g., check that output files are non-empty and contain expected markdown structure before reporting success.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~300+ lines. It includes unnecessary sections like 'Visual Enhancement with Scientific Schematics' (irrelevant to the core skill), explains obvious concepts ('Markdown is token-efficient and well-understood by modern language models'), repeats similar code patterns across 6+ use cases that are trivially different, and includes a 'Key Benefits' marketing-style list. Many examples are near-duplicates (e.g., batch convert appears twice with minor variations). | 1 / 3 |
Actionability | The skill provides fully executable, copy-paste ready code examples for CLI usage, Python API, streaming, AI-enhanced conversion, batch processing, and error handling. Commands and code are concrete and complete with proper imports and context. | 3 / 3 |
Workflow Clarity | Individual operations are clear, but there's no cohesive workflow with validation checkpoints. The batch processing examples lack verification that conversions succeeded beyond catching exceptions. There's no guidance on validating output quality or handling partial failures in batch operations. | 2 / 3 |
Progressive Disclosure | The document references external files (references/api_reference.md, references/file_formats.md, scripts/) in the 'Next Steps' section, which is good. However, the main body is a monolithic wall containing extensive inline content (6 use cases, troubleshooting, performance notes, integration examples) that should be split into separate reference files. The supported formats table and all the use case examples bloat the overview. | 2 / 3 |
Total | 8 / 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
b58ad7e
- 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.