Content
42%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 provides a reasonable reference for UniProt REST API access but suffers from verbosity (explaining what UniProt is, restating capabilities multiple times) and incomplete actionability (query strings shown but no complete HTTP request examples). The workflow for ID mapping is outlined but lacks error handling/validation steps, and the referenced helper script and supporting files are not provided in the bundle.
Suggestions
Remove the overview paragraph explaining what UniProt is and the 'When to Use This Skill' section — Claude already knows these things. Start directly with Core Capabilities.
Add a complete, executable Python example using the requests library that performs a search and parses the JSON response, rather than just showing URL patterns and query strings.
Add error handling and validation steps to the ID mapping workflow (e.g., check for failed jobs, validate input IDs, handle rate limiting responses with retry logic).
Move the detailed query syntax examples to the referenced query_syntax.md file and keep only 1-2 key examples inline to reduce the main file length.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The overview section explains what UniProt is (Claude already knows this), the 'When to Use This Skill' section is an 8-item list that largely restates the overview, and there's significant redundancy throughout. The best practices section includes obvious advice like 'choose the most appropriate format' and 'cache results.' Much of this could be cut in half. | 2 / 5 |
Actionability | The skill provides URL patterns and query syntax examples which are useful, but the Python implementation section references a helper script (`scripts/uniprot_client.py`) without showing its actual code. The query examples are illustrative but not fully executable end-to-end — there's no complete working example showing how to make an HTTP request and parse the response. The code blocks show query strings but not the actual requests library calls. | 3 / 5 |
Workflow Clarity | The ID mapping workflow has a clear 3-step sequence (submit, check status, retrieve results), but lacks validation/error handling steps — what if the job fails? What if IDs are invalid? There's no feedback loop for error recovery. The general search workflow is implicit rather than explicitly sequenced. For batch operations with up to 100,000 IDs, the absence of validation checkpoints is notable. | 3 / 5 |
Progressive Disclosure | The skill references several supporting files (api_fields.md, id_mapping_databases.md, query_syntax.md, api_examples.md, uniprot_client.py) which is good structure, but no bundle files were provided, so these references are unverifiable. The main file itself is quite long (~180 lines) and inlines content that could be in reference files (e.g., the full query syntax examples section could live in the referenced query_syntax.md). The references are reasonably well-signaled though. | 3 / 5 |
Total | 11 / 20 Passed |