Content
67%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
Well-structured, lean content with a clear numbered workflow and useful tool-surface table. The main defect is a malformed Authentication code block that breaks copy-paste usability of a core section.
Suggestions
Fix the Authentication code block: close the ``` fence and separate the token placeholder from the '401 Unauthorized' sentence so the Bearer header example is copy-paste ready.
Consider moving the 'Implementation & maintenance (for Space developers)' section into a separate reference file to keep SKILL.md focused on agent usage and improve conciseness.
Add a brief example request/response or a sample search_plugins invocation so the actionability reaches fully executable coverage.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body assumes Claude knows MCP, streamable HTTP, and Bearer auth and stays lean, but the "Implementation & maintenance" section is developer-focused material that adds tokens beyond what an agent user needs. | 4 / 5 |
Actionability | It provides a concrete endpoint, client-config JSON, a tool table, and numbered steps, but the Authentication code block is malformed (missing closing fence, "lbpat_...uests without a valid PAT" run together), making that core example non-executable. | 3 / 5 |
Workflow Clarity | The numbered "How to use" steps (1-4) give a clear sequence for a read-only browse/search flow; no validation checkpoints are needed since the server is explicitly read-only, so the destructive/batch cap does not apply. | 4 / 5 |
Progressive Disclosure | Sections (Endpoint, Authentication, Client configuration, Tool surface, How to use, Pitfalls) are well-organized with one-level-deep developer file references; at ~80 lines with the tool table and implementation notes inlined it stops short of a clean 5. | 4 / 5 |
Total | 15 / 20 Passed |