Content
65%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.
The skill is highly actionable with copy-paste-ready commands across many MCP scenarios, but it is over-long with repeated examples and inlines reference material that would benefit from separate files. Destructive workflows lack explicit validation checkpoints.
Suggestions
Move the server catalog, transport, and authentication detail into reference files (e.g., SERVERS.md, AUTH.md) and link to them from a concise overview to improve progressive disclosure.
Add explicit validate-then-proceed checkpoints to the Complete Workflow for destructive calls (e.g., verify a write_file succeeded by reading the file back before continuing).
De-duplicate examples that recur between Discovery, Common MCP Servers, and Best Practices to tighten token use.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly executable commands with little concept over-explanation, but several sections repeat earlier examples (e.g., Common MCP Servers re-lists filesystem/memory/github commands already shown in Discovery) and could be tightened. | 3 / 5 |
Actionability | Commands are fully executable and copy-paste ready, with concrete examples spanning filesystem, memory, HTTP, and auth cases covering the common usage. | 5 / 5 |
Workflow Clarity | Discovery steps and the Complete Workflow are sequenced, but workflows involving destructive tool calls (write_file, delete) lack explicit validate-after-call checkpoints; the destructive/batch cap limits this to 3 despite a present error-handling best practice. | 3 / 5 |
Progressive Disclosure | Content is well-sectioned with a Quick Reference table, but ~370 lines of reference material (server catalog, auth, transports) are inlined in SKILL.md with no bundle files to split deeper detail into; the under-50-line exception does not apply. | 3 / 5 |
Total | 14 / 20 Passed |